diff --git a/records/track_10min_16mb/2026-03-26_10L_NgramBackoff_MatrixLR03/README.md b/records/track_10min_16mb/2026-03-26_10L_NgramBackoff_MatrixLR03/README.md new file mode 100644 index 000000000..90f9652d1 --- /dev/null +++ b/records/track_10min_16mb/2026-03-26_10L_NgramBackoff_MatrixLR03/README.md @@ -0,0 +1,46 @@ +# Record: 10L + Multi-Order N-gram Backoff + Matrix LR 0.03 + +**val_bpb = 0.9076** (seed 42, additional seeds pending) | **15.32 MB** | 8xH100 SXM, 600s + +## Results + +| Seed | Steps | ms/step | Pre-quant BPB | **N-gram BPB** | Artifact | +|------|-------|---------|---------------|----------------|----------| +| 42 | 6,693 | 89.6 | 1.1528 | **0.9076** | 15,320,749 | + +## Key Change from PR #802 + +Single change: **MATRIX_LR=0.03** (from 0.02). This was discovered through systematic RTX4500 screening (20 experiments) to be the single largest training hyperparameter improvement for 10L architectures (-0.064 BPB on RTX4500 screening, -0.005 BPB on 8xH100 full run). + +## Architecture (same as PR #802) + +- 10L, 512d, GQA 8H/4KV, MLP 3x LeakyReLU(0.5)² +- BigramHash(4096, dim=128), SmearGate, Value Residual, Gated Attention +- XSA last 4 layers, Partial RoPE 16/64, LN Scale +- U-Net skip connections, tied embeddings, logit softcap=30 + +## Training + +- Muon optimizer: **lr=0.03** (was 0.02), momentum 0.92→0.99, WD=0.04 +- EMA(0.997), warmdown=3500 steps +- Mixed int5-MLP/int6-attn quantization + zstd-22 +- 3% magnitude pruning + +## Eval: Multi-Order N-gram Backoff (from PR #802) + +- Score-first backward-looking n-gram cache (orders 2-7) +- Highest matching order wins (backoff from 7-gram to bigram) +- Entropy-adaptive alpha: `alpha = 0.05 + 0.55 * sigmoid(2 * (H - 4.0))` +- 4M XOR-hash buckets, min_count=2 +- **Legal:** each token scored BEFORE cache is updated + +## Reproduction + +```bash +MATRIX_LR=0.03 torchrun --standalone --nproc_per_node=8 train_gpt.py +``` + +## Based On + +- PR #802 (@[author]): 10L + Multi-Order N-gram Backoff (0.9123 BPB) +- Our systematic hyperparameter screening (steps 10-12, 74 experiments) diff --git a/records/track_10min_16mb/2026-03-26_10L_NgramBackoff_MatrixLR03/submission.json b/records/track_10min_16mb/2026-03-26_10L_NgramBackoff_MatrixLR03/submission.json new file mode 100644 index 000000000..aa2d39a45 --- /dev/null +++ b/records/track_10min_16mb/2026-03-26_10L_NgramBackoff_MatrixLR03/submission.json @@ -0,0 +1,13 @@ +{ + "track": "10min_16mb", + "date": "2026-03-26", + "name": "Record: 10L + Multi-Order N-gram Backoff + Matrix LR 0.03 (val_bpb=0.9076)", + "author": "bigbag", + "github": "bigbag", + "seed_results": { + "42": {"val_loss": 1.53248692, "val_bpb": 0.90762747, "artifact_bytes": 15320749} + }, + "mean_val_loss": 1.53248692, + "mean_val_bpb": 0.90762747, + "code_bytes": 68444 +} diff --git a/records/track_10min_16mb/2026-03-26_10L_NgramBackoff_MatrixLR03/train.log b/records/track_10min_16mb/2026-03-26_10L_NgramBackoff_MatrixLR03/train.log new file mode 100644 index 000000000..45c7bc48c --- /dev/null +++ b/records/track_10min_16mb/2026-03-26_10L_NgramBackoff_MatrixLR03/train.log @@ -0,0 +1,119 @@ +W0326 07:28:29.013000 143 torch/distributed/run.py:803] +W0326 07:28:29.013000 143 torch/distributed/run.py:803] ***************************************** +W0326 07:28:29.013000 143 torch/distributed/run.py:803] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. +W0326 07:28:29.013000 143 torch/distributed/run.py:803] ***************************************** +logs/400c6056-3073-4716-ab99-5e1a14a95d10.txt +val_bpb:enabled tokenizer_kind=sentencepiece tokenizer_path=./data/tokenizers/fineweb_1024_bpe.model +train_loader:dataset:fineweb10B_sp1024 train_shards:80 +val_loader:shards pattern=./data/datasets/fineweb10B_sp1024/fineweb_val_*.bin tokens:62021632 +model_params:24730705 +world_size:8 grad_accum_steps:1 +attention_mode:gqa num_heads:8 num_kv_heads:4 +tie_embeddings:True embed_lr:0.03 matrix_lr:0.03 scalar_lr:0.02 +train_batch_tokens:786432 train_seq_len:2048 iterations:20000 warmup_steps:5 max_wallclock_seconds:600.000 +seed:42 +warmup_step:1/5 +warmup_step:2/5 +warmup_step:3/5 +warmup_step:4/5 +warmup_step:5/5 +step:1/20000 train_loss:6.9311 train_time:147ms step_avg:147.24ms +step:2/20000 train_loss:7.9786 train_time:232ms step_avg:116.01ms +step:3/20000 train_loss:7.3922 train_time:319ms step_avg:106.37ms +step:4/20000 train_loss:7.0074 train_time:407ms step_avg:101.75ms +step:5/20000 train_loss:6.8978 train_time:495ms step_avg:98.91ms +step:6/20000 train_loss:6.8527 train_time:583ms step_avg:97.09ms +step:7/20000 train_loss:6.7423 train_time:670ms step_avg:95.76ms +step:8/20000 train_loss:6.7231 train_time:758ms step_avg:94.79ms +step:9/20000 train_loss:6.3524 train_time:847ms step_avg:94.09ms +step:10/20000 train_loss:6.0305 train_time:945ms step_avg:94.52ms +step:100/20000 train_loss:3.1302 train_time:8864ms step_avg:88.64ms +step:200/20000 train_loss:2.3178 train_time:17762ms step_avg:88.81ms +step:300/20000 train_loss:2.5089 train_time:26701ms step_avg:89.00ms +step:400/20000 train_loss:2.3906 train_time:35653ms step_avg:89.13ms +step:500/20000 train_loss:2.3830 train_time:44548ms step_avg:89.10ms +step:600/20000 train_loss:2.3275 train_time:53505ms step_avg:89.18ms +step:700/20000 train_loss:2.3416 train_time:62473ms step_avg:89.25ms +step:800/20000 train_loss:2.2354 train_time:71439ms step_avg:89.30ms +step:900/20000 train_loss:2.1295 train_time:80397ms step_avg:89.33ms +step:1000/20000 train_loss:2.2768 train_time:89292ms step_avg:89.29ms +step:1100/20000 train_loss:2.3216 train_time:98263ms step_avg:89.33ms +step:1200/20000 train_loss:2.3596 train_time:107250ms step_avg:89.38ms +step:1300/20000 train_loss:2.1052 train_time:116228ms step_avg:89.41ms +step:1400/20000 train_loss:2.1903 train_time:125214ms step_avg:89.44ms +step:1500/20000 train_loss:2.2287 train_time:134126ms step_avg:89.42ms +step:1600/20000 train_loss:2.0827 train_time:143110ms step_avg:89.44ms +step:1700/20000 train_loss:2.1530 train_time:152101ms step_avg:89.47ms +step:1800/20000 train_loss:2.1613 train_time:161091ms step_avg:89.49ms +step:1900/20000 train_loss:2.1324 train_time:170015ms step_avg:89.48ms +step:2000/20000 train_loss:2.0775 train_time:179010ms step_avg:89.51ms +step:2100/20000 train_loss:2.0585 train_time:188003ms step_avg:89.53ms +step:2200/20000 train_loss:2.1790 train_time:196990ms step_avg:89.54ms +step:2300/20000 train_loss:2.1175 train_time:205986ms step_avg:89.56ms +step:2400/20000 train_loss:2.0752 train_time:214900ms step_avg:89.54ms +step:2500/20000 train_loss:2.1815 train_time:223884ms step_avg:89.55ms +step:2600/20000 train_loss:2.1219 train_time:232872ms step_avg:89.57ms +step:2700/20000 train_loss:2.1168 train_time:241849ms step_avg:89.57ms +step:2800/20000 train_loss:2.1687 train_time:250835ms step_avg:89.58ms +step:2900/20000 train_loss:2.0388 train_time:259753ms step_avg:89.57ms +step:3000/20000 train_loss:2.1761 train_time:268744ms step_avg:89.58ms +step:3100/20000 train_loss:2.0509 train_time:277729ms step_avg:89.59ms +step:3200/20000 train_loss:2.1889 train_time:286723ms step_avg:89.60ms +step:3300/20000 train_loss:2.0859 train_time:295643ms step_avg:89.59ms +step:3400/20000 train_loss:2.0372 train_time:304639ms step_avg:89.60ms +step:3500/20000 train_loss:2.1929 train_time:313626ms step_avg:89.61ms +step:3600/20000 train_loss:2.1083 train_time:322610ms step_avg:89.61ms +step:3700/20000 train_loss:2.1014 train_time:331604ms step_avg:89.62ms +step:3800/20000 train_loss:2.0793 train_time:340517ms step_avg:89.61ms +step:3900/20000 train_loss:2.0826 train_time:349506ms step_avg:89.62ms +step:4000/20000 train_loss:1.9847 train_time:358487ms step_avg:89.62ms +step:4100/20000 train_loss:2.0224 train_time:367477ms step_avg:89.63ms +step:4200/20000 train_loss:2.1579 train_time:376471ms step_avg:89.64ms +step:4300/20000 train_loss:2.0687 train_time:385386ms step_avg:89.62ms +step:4400/20000 train_loss:2.0379 train_time:394370ms step_avg:89.63ms +step:4500/20000 train_loss:2.1302 train_time:403353ms step_avg:89.63ms +step:4600/20000 train_loss:1.8487 train_time:412338ms step_avg:89.64ms +step:4700/20000 train_loss:2.2395 train_time:421258ms step_avg:89.63ms +step:4800/20000 train_loss:2.4318 train_time:430246ms step_avg:89.63ms +step:4900/20000 train_loss:2.0563 train_time:439232ms step_avg:89.64ms +step:5000/20000 train_loss:2.1108 train_time:448222ms step_avg:89.64ms +step:5100/20000 train_loss:2.1305 train_time:457207ms step_avg:89.65ms +step:5200/20000 train_loss:2.0503 train_time:466122ms step_avg:89.64ms +step:5300/20000 train_loss:2.0182 train_time:475110ms step_avg:89.64ms +step:5400/20000 train_loss:2.0540 train_time:484085ms step_avg:89.65ms +step:5500/20000 train_loss:2.0202 train_time:493054ms step_avg:89.65ms +step:5600/20000 train_loss:1.9588 train_time:502041ms step_avg:89.65ms +step:5700/20000 train_loss:2.0162 train_time:510954ms step_avg:89.64ms +step:5800/20000 train_loss:2.0034 train_time:519939ms step_avg:89.64ms +step:5900/20000 train_loss:1.9065 train_time:528922ms step_avg:89.65ms +step:6000/20000 train_loss:1.9426 train_time:537914ms step_avg:89.65ms +step:6100/20000 train_loss:1.9187 train_time:546829ms step_avg:89.64ms +step:6200/20000 train_loss:1.9524 train_time:555811ms step_avg:89.65ms +step:6300/20000 train_loss:1.9471 train_time:564800ms step_avg:89.65ms +step:6400/20000 train_loss:1.9972 train_time:573786ms step_avg:89.65ms +step:6500/20000 train_loss:2.0796 train_time:582758ms step_avg:89.66ms +step:6600/20000 train_loss:1.8403 train_time:591663ms step_avg:89.65ms +step:6693/20000 val_loss:1.9464 val_bpb:1.1528 train_time:600024ms step_avg:89.65ms +stopping_early: wallclock_cap train_time:600024ms step:6693/20000 +peak memory allocated: 20944 MiB reserved: 21086 MiB +ema:applying shadow model +Serialized model: 96864555 bytes +Code size: 68444 bytes +Total submission size: 96932999 bytes +Serialized model int6+zstd: 15252305 bytes +Total submission size: 15320749 bytes (15.32 MB) +SIZE CHECK PASSED: 15.32 MB < 16.00 MB +final_eval_mode:sliding_ngram orders=2-7 alpha=0.4 entropy=True stride:64 +ngram_cache:enabled orders=2-7 backoff entropy=True alpha=0.4 ent_base=0.05 ent_range=0.55 min_count=2 buckets=4194304 + ngram_eval [ 10.6%] bpb=1.110706 t=23s + ngram_eval [ 21.2%] bpb=1.092910 t=35s + ngram_eval [ 31.8%] bpb=1.068654 t=47s + ngram_eval [ 42.3%] bpb=1.039567 t=59s + ngram_eval [ 52.9%] bpb=1.011695 t=71s + ngram_eval [ 63.5%] bpb=0.985400 t=84s + ngram_eval [ 74.0%] bpb=0.963914 t=96s + ngram_eval [ 84.6%] bpb=0.942984 t=108s + ngram_eval [ 95.2%] bpb=0.922574 t=120s + ngram_eval DONE: bpb=0.907627 tokens=62023616 t=138s +final_int8_zlib_roundtrip val_loss:1.5325 val_bpb:0.9076 eval_time:138129ms +final_int8_zlib_roundtrip_exact val_loss:1.53248692 val_bpb:0.90762747 diff --git a/records/track_10min_16mb/2026-03-26_10L_NgramBackoff_MatrixLR03/train_gpt.py b/records/track_10min_16mb/2026-03-26_10L_NgramBackoff_MatrixLR03/train_gpt.py new file mode 100644 index 000000000..7721b5a33 --- /dev/null +++ b/records/track_10min_16mb/2026-03-26_10L_NgramBackoff_MatrixLR03/train_gpt.py @@ -0,0 +1,1541 @@ +""" +The `train_gpt.py` and `train_gpt_mlx.py` scripts are intended as good launching-off points for new participants, not SOTA configs. We'll accept PRs that tune, improve, or simplify these scripts without significantly increasing complexity, but competitive submissions should stay in the `/records` folder. + +Hard stop: To keep readable for newcomers, let's make sure `train_gpt.py` and `train_gpt_mlx.py` never are longer than 1500 lines. +""" + +from __future__ import annotations + +import copy +import glob +import io +import math +import os +import random +import subprocess +import sys +import time +import uuid +import zlib +from pathlib import Path + +try: + import zstandard + _COMPRESSOR = "zstd" +except ImportError: + _COMPRESSOR = "zlib" + +import numpy as np +import sentencepiece as spm +import torch +import torch.distributed as dist +import torch.nn.functional as F +from torch import Tensor, nn +from torch.nn.parallel import DistributedDataParallel as DDP + +# ----------------------------- +# HYPERPARAMETERS +# ----------------------------- + +class Hyperparameters: + data_path = os.environ.get("DATA_PATH", "./data/datasets/fineweb10B_sp1024") + train_files = os.path.join(data_path, "fineweb_train_*.bin") + val_files = os.path.join(data_path, "fineweb_val_*.bin") + tokenizer_path = os.environ.get("TOKENIZER_PATH", "./data/tokenizers/fineweb_1024_bpe.model") + run_id = os.environ.get("RUN_ID", str(uuid.uuid4())) + seed = int(os.environ.get("SEED", 42)) + + val_batch_size = int(os.environ.get("VAL_BATCH_SIZE", 524_288)) + val_loss_every = int(os.environ.get("VAL_LOSS_EVERY", 0)) # 0=skip mid-train val, maximize training steps + train_log_every = int(os.environ.get("TRAIN_LOG_EVERY", 100)) + + iterations = int(os.environ.get("ITERATIONS", 20000)) + warmdown_iters = int(os.environ.get("WARMDOWN_ITERS", 3500)) + warmup_steps = int(os.environ.get("WARMUP_STEPS", 5)) # minimal warmup, maximize real steps + train_batch_tokens = int(os.environ.get("TRAIN_BATCH_TOKENS", 786_432)) + train_seq_len = int(os.environ.get("TRAIN_SEQ_LEN", 2048)) + max_wallclock_seconds = float(os.environ.get("MAX_WALLCLOCK_SECONDS", 600.0)) + qk_gain_init = float(os.environ.get("QK_GAIN_INIT", 1.5)) + + vocab_size = int(os.environ.get("VOCAB_SIZE", 1024)) + num_layers = int(os.environ.get("NUM_LAYERS", 10)) + num_kv_heads = int(os.environ.get("NUM_KV_HEADS", 4)) + model_dim = int(os.environ.get("MODEL_DIM", 512)) + num_heads = int(os.environ.get("NUM_HEADS", 8)) + mlp_mult = float(os.environ.get("MLP_MULT", 3.0)) + tie_embeddings = bool(int(os.environ.get("TIE_EMBEDDINGS", "1"))) + rope_base = float(os.environ.get("ROPE_BASE", 10000.0)) + logit_softcap = float(os.environ.get("LOGIT_SOFTCAP", 30.0)) + + embed_lr = float(os.environ.get("EMBED_LR", 0.6)) + head_lr = float(os.environ.get("HEAD_LR", 0.008)) + tied_embed_lr = float(os.environ.get("TIED_EMBED_LR", 0.03)) + tied_embed_init_std = float(os.environ.get("TIED_EMBED_INIT_STD", 0.005)) + matrix_lr = float(os.environ.get("MATRIX_LR", 0.02)) + scalar_lr = float(os.environ.get("SCALAR_LR", 0.02)) + muon_momentum = float(os.environ.get("MUON_MOMENTUM", 0.99)) + muon_backend_steps = int(os.environ.get("MUON_BACKEND_STEPS", 5)) + muon_momentum_warmup_start = float(os.environ.get("MUON_MOMENTUM_WARMUP_START", 0.92)) + muon_momentum_warmup_steps = int(os.environ.get("MUON_MOMENTUM_WARMUP_STEPS", 1500)) + beta1 = float(os.environ.get("BETA1", 0.9)) + beta2 = float(os.environ.get("BETA2", 0.95)) + adam_eps = float(os.environ.get("ADAM_EPS", 1e-8)) + grad_clip_norm = float(os.environ.get("GRAD_CLIP_NORM", 0.3)) + weight_decay = float(os.environ.get("WEIGHT_DECAY", 0.04)) + + eval_stride = int(os.environ.get("EVAL_STRIDE", 64)) + eval_batch_seqs = int(os.environ.get("EVAL_BATCH_SEQS", 64)) # larger batch for faster eval (no gradients) + + bigram_vocab_size = int(os.environ.get("BIGRAM_VOCAB_SIZE", 4096)) + bigram_dim = int(os.environ.get("BIGRAM_DIM", 128)) + + # Partial RoPE: only rotate first rope_dims dims (0 = full head_dim) + rope_dims = int(os.environ.get("ROPE_DIMS", 16)) + + # LN Scale: dampen norm inputs by 1/sqrt(layer_idx+1) for deeper layers + ln_scale = bool(int(os.environ.get("LN_SCALE", "1"))) + + # XSA: exclusive self-attention on last N layers (0 = disabled) + xsa_last_n = int(os.environ.get("XSA_LAST_N", 4)) # proven: last 4 layers + + # EMA: exponential moving average (replaces SWA when enabled) + ema_enabled = bool(int(os.environ.get("EMA_ENABLED", "1"))) + ema_decay = float(os.environ.get("EMA_DECAY", 0.997)) + + swa_enabled = bool(int(os.environ.get("SWA_ENABLED", "0"))) # OFF by default, EMA replaces it + swa_start_frac = float(os.environ.get("SWA_START_FRAC", 0.4)) + swa_every = int(os.environ.get("SWA_EVERY", 50)) + + # N-gram eval cache: multi-order backoff + entropy-adaptive alpha (score-first, legal) + ngram_eval_max_order = int(os.environ.get("NGRAM_EVAL_ORDER", 7)) # max n-gram order + ngram_eval_min_order = int(os.environ.get("NGRAM_EVAL_MIN_ORDER", 2)) # min backoff order + ngram_eval_alpha = float(os.environ.get("NGRAM_EVAL_ALPHA", 0.40)) # base alpha + ngram_eval_min_count = int(os.environ.get("NGRAM_EVAL_MIN_COUNT", 2)) + ngram_eval_buckets = int(os.environ.get("NGRAM_EVAL_BUCKETS", 4_194_304)) + ngram_eval_entropy = bool(int(os.environ.get("NGRAM_EVAL_ENTROPY", "1"))) + ngram_eval_ent_base = float(os.environ.get("NGRAM_EVAL_ENT_BASE", 0.05)) + ngram_eval_ent_range = float(os.environ.get("NGRAM_EVAL_ENT_RANGE", 0.55)) + ngram_eval_ent_scale = float(os.environ.get("NGRAM_EVAL_ENT_SCALE", 2.0)) + ngram_eval_ent_thresh = float(os.environ.get("NGRAM_EVAL_ENT_THRESH", 4.0)) + +# ----------------------------- +# MUON OPTIMIZER +# ----------------------------- + +def zeropower_via_newtonschulz5(G: Tensor, steps: int = 10, eps: float = 1e-7) -> Tensor: + a, b, c = (3.4445, -4.7750, 2.0315) + X = G.bfloat16() + X /= X.norm() + eps + transposed = G.size(0) > G.size(1) + if transposed: + X = X.T + for _ in range(steps): + A = X @ X.T + B = b * A + c * A @ A + X = a * X + B @ X + return X.T if transposed else X + + +class Muon(torch.optim.Optimizer): + def __init__(self, params, lr: float, momentum: float, backend_steps: int, nesterov: bool = True, weight_decay: float = 0.0): + super().__init__( + params, + dict(lr=lr, momentum=momentum, backend_steps=backend_steps, nesterov=nesterov, weight_decay=weight_decay), + ) + + @torch.no_grad() + def step(self, closure=None): + loss = None + if closure is not None: + with torch.enable_grad(): + loss = closure() + distributed = dist.is_available() and dist.is_initialized() + world_size = dist.get_world_size() if distributed else 1 + rank = dist.get_rank() if distributed else 0 + + for group in self.param_groups: + params = group["params"] + if not params: + continue + lr = group["lr"] + momentum = group["momentum"] + backend_steps = group["backend_steps"] + nesterov = group["nesterov"] + + total_params = sum(int(p.numel()) for p in params) + updates_flat = torch.zeros(total_params, device=params[0].device, dtype=torch.bfloat16) + + curr = 0 + for i, p in enumerate(params): + if i % world_size == rank and p.grad is not None: + g = p.grad + state = self.state[p] + if "momentum_buffer" not in state: + state["momentum_buffer"] = torch.zeros_like(g) + buf = state["momentum_buffer"] + buf.mul_(momentum).add_(g) + if nesterov: + g = g.add(buf, alpha=momentum) + g = zeropower_via_newtonschulz5(g, steps=backend_steps) + g *= max(1, g.size(0) / g.size(1)) ** 0.5 + updates_flat[curr : curr + p.numel()] = g.reshape(-1) + curr += p.numel() + + if distributed: + dist.all_reduce(updates_flat, op=dist.ReduceOp.SUM) + + wd = group.get("weight_decay", 0.0) + curr = 0 + for p in params: + g = updates_flat[curr : curr + p.numel()].view_as(p).to(dtype=p.dtype) + if wd > 0: + p.data.mul_(1.0 - lr * wd) + p.add_(g, alpha=-lr) + curr += p.numel() + return loss + + +# ----------------------------- +# TOKENIZER-AGNOSTIC EVALUATION +# ----------------------------- + +def build_sentencepiece_luts( + sp: spm.SentencePieceProcessor, vocab_size: int, device: torch.device +) -> tuple[Tensor, Tensor, Tensor]: + sp_vocab_size = int(sp.vocab_size()) + table_size = max(sp_vocab_size, vocab_size) + base_bytes_np = np.zeros((table_size,), dtype=np.int16) + has_leading_space_np = np.zeros((table_size,), dtype=np.bool_) + is_boundary_token_np = np.ones((table_size,), dtype=np.bool_) + for token_id in range(sp_vocab_size): + if sp.is_control(token_id) or sp.is_unknown(token_id) or sp.is_unused(token_id): + continue + is_boundary_token_np[token_id] = False + if sp.is_byte(token_id): + base_bytes_np[token_id] = 1 + continue + piece = sp.id_to_piece(token_id) + if piece.startswith("\u2581"): + has_leading_space_np[token_id] = True + piece = piece[1:] + base_bytes_np[token_id] = len(piece.encode("utf-8")) + return ( + torch.tensor(base_bytes_np, dtype=torch.int16, device=device), + torch.tensor(has_leading_space_np, dtype=torch.bool, device=device), + torch.tensor(is_boundary_token_np, dtype=torch.bool, device=device), + ) + + +def load_validation_tokens(pattern: str, seq_len: int) -> Tensor: + files = [Path(p) for p in sorted(glob.glob(pattern))] + if not files: + raise FileNotFoundError(f"No files found for pattern: {pattern}") + tokens = torch.cat([load_data_shard(file) for file in files]).contiguous() + usable = ((tokens.numel() - 1) // seq_len) * seq_len + if usable <= 0: + raise ValueError(f"Validation split is too short for TRAIN_SEQ_LEN={seq_len}") + return tokens[: usable + 1] + + +def eval_val( + args: Hyperparameters, + model: nn.Module, + rank: int, + world_size: int, + device: torch.device, + grad_accum_steps: int, + val_tokens: Tensor, + base_bytes_lut: Tensor, + has_leading_space_lut: Tensor, + is_boundary_token_lut: Tensor, +) -> tuple[float, float]: + local_batch_tokens = args.val_batch_size // (world_size * grad_accum_steps) + if local_batch_tokens < args.train_seq_len: + raise ValueError( + "VAL_BATCH_SIZE must provide at least one sequence per rank; " + f"got VAL_BATCH_SIZE={args.val_batch_size}, WORLD_SIZE={world_size}, " + f"GRAD_ACCUM_STEPS={grad_accum_steps}, TRAIN_SEQ_LEN={args.train_seq_len}" + ) + local_batch_seqs = local_batch_tokens // args.train_seq_len + total_seqs = (val_tokens.numel() - 1) // args.train_seq_len + seq_start = (total_seqs * rank) // world_size + seq_end = (total_seqs * (rank + 1)) // world_size + val_loss_sum = torch.zeros((), device=device, dtype=torch.float64) + val_token_count = torch.zeros((), device=device, dtype=torch.float64) + val_byte_count = torch.zeros((), device=device, dtype=torch.float64) + model.eval() + with torch.inference_mode(): + for batch_seq_start in range(seq_start, seq_end, local_batch_seqs): + batch_seq_end = min(batch_seq_start + local_batch_seqs, seq_end) + raw_start = batch_seq_start * args.train_seq_len + raw_end = batch_seq_end * args.train_seq_len + 1 + local = val_tokens[raw_start:raw_end].to(device=device, dtype=torch.int64, non_blocking=True) + x = local[:-1].reshape(-1, args.train_seq_len) + y = local[1:].reshape(-1, args.train_seq_len) + with torch.autocast(device_type="cuda", dtype=torch.bfloat16, enabled=True): + batch_loss = model(x, y).detach() + batch_token_count = float(y.numel()) + val_loss_sum += batch_loss.to(torch.float64) * batch_token_count + val_token_count += batch_token_count + prev_ids = x.reshape(-1) + tgt_ids = y.reshape(-1) + token_bytes = base_bytes_lut[tgt_ids].to(dtype=torch.int16) + token_bytes += (has_leading_space_lut[tgt_ids] & ~is_boundary_token_lut[prev_ids]).to(dtype=torch.int16) + val_byte_count += token_bytes.to(torch.float64).sum() + if dist.is_available() and dist.is_initialized(): + dist.all_reduce(val_loss_sum, op=dist.ReduceOp.SUM) + dist.all_reduce(val_token_count, op=dist.ReduceOp.SUM) + dist.all_reduce(val_byte_count, op=dist.ReduceOp.SUM) + val_loss = val_loss_sum / val_token_count + bits_per_token = val_loss.item() / math.log(2.0) + tokens_per_byte = val_token_count.item() / val_byte_count.item() + model.train() + return float(val_loss.item()), float(bits_per_token * tokens_per_byte) + + +# ----------------------------- +# POST-TRAINING QUANTIZATION (INT8 legacy + INT6 mixed) +# ----------------------------- + +CONTROL_TENSOR_NAME_PATTERNS = tuple( + pattern + for pattern in os.environ.get( + "CONTROL_TENSOR_NAME_PATTERNS", + "attn_scale,attn_scales,mlp_scale,mlp_scales,resid_mix,resid_mixes,q_gain,skip_weight,skip_weights,smear,bigram.scale", + ).split(",") + if pattern +) +FP16_KEEP_NAME_PATTERNS = tuple( + pattern + for pattern in os.environ.get("FP16_KEEP_NAME_PATTERNS", "tok_emb,blocks.8.attn.c_k").split(",") + if pattern +) +INT8_KEEP_FLOAT_FP32_NAME_PATTERNS = tuple( + pattern + for pattern in os.environ.get( + "INT8_KEEP_FLOAT_FP32_NAME_PATTERNS", + ",".join(CONTROL_TENSOR_NAME_PATTERNS), + ).split(",") + if pattern +) +INT8_KEEP_FLOAT_MAX_NUMEL = 65_536 +INT8_KEEP_FLOAT_STORE_DTYPE = torch.float16 +INT8_PER_ROW_SCALE_DTYPE = torch.float16 +INT8_CLIP_PERCENTILE = 99.99984 +INT8_CLIP_Q = INT8_CLIP_PERCENTILE / 100.0 + +def tensor_nbytes(t: Tensor) -> int: + return int(t.numel()) * int(t.element_size()) + +def quantize_float_tensor(t: Tensor) -> tuple[Tensor, Tensor]: + t32 = t.float() + if t32.ndim == 2: + clip_abs = ( + torch.quantile(t32.abs(), INT8_CLIP_Q, dim=1) + if t32.numel() + else torch.empty((t32.shape[0],), dtype=torch.float32) + ) + clipped = torch.maximum(torch.minimum(t32, clip_abs[:, None]), -clip_abs[:, None]) + scale = (clip_abs / 127.0).clamp_min(1.0 / 127.0) + q = torch.clamp(torch.round(clipped / scale[:, None]), -127, 127).to(torch.int8).contiguous() + return q, scale.to(dtype=INT8_PER_ROW_SCALE_DTYPE).contiguous() + clip_abs = float(torch.quantile(t32.abs().flatten(), INT8_CLIP_Q).item()) if t32.numel() else 0.0 + scale = torch.tensor(clip_abs / 127.0 if clip_abs > 0 else 1.0, dtype=torch.float32) + q = torch.clamp(torch.round(torch.clamp(t32, -clip_abs, clip_abs) / scale), -127, 127).to(torch.int8).contiguous() + return q, scale + + +def _classify_param(name: str) -> str: + if "tok_emb" in name or "lm_head" in name: + return "embed" + if ".mlp." in name: + return "mlp" + if "bigram" in name: + return "bigram" + if ".attn." in name or (".proj." in name and ".mlp." not in name): + return "attn" + return "other" + +def quantize_intN_per_row(t: Tensor, clip_range: int = 31) -> tuple[Tensor, Tensor]: + t32 = t.float() + if t32.ndim == 2: + row_max = t32.abs().amax(dim=1) + scale = (row_max / clip_range).clamp_min(1e-12).to(torch.float16) + scale = scale.clamp_min(torch.finfo(torch.float16).tiny) + q = torch.clamp(torch.round(t32 / scale.float()[:, None]), -(clip_range+1), clip_range).to(torch.int8) + return q, scale + amax = t32.abs().max().item() + scale = torch.tensor(max(amax / clip_range, 1e-12), dtype=torch.float16) + q = torch.clamp(torch.round(t32 / scale.float()), -(clip_range+1), clip_range).to(torch.int8) + return q, scale + +def mixed_quantize_int6(state_dict: dict[str, Tensor], int6_cats: set[str]): + result: dict[str, Tensor] = {} + meta: dict[str, object] = {} + for name, tensor in state_dict.items(): + t = tensor.detach().cpu().contiguous() + cat = _classify_param(name) + if not t.is_floating_point() or t.numel() <= 8192: + result[name] = t.to(torch.float16) if t.is_floating_point() else t + meta[name] = "passthrough" + continue + if any(p in name for p in CONTROL_TENSOR_NAME_PATTERNS): + result[name] = t.float() + meta[name] = "passthrough_ctrl" + continue + if any(pattern in name for pattern in FP16_KEEP_NAME_PATTERNS): + result[name] = t.to(dtype=torch.float16).contiguous() + meta[name] = "passthrough_fp16" + continue + if cat in int6_cats and t.ndim >= 1: + clip = 15 if cat == "mlp" else 31 # int5 for MLP, int6 for attention + q, s = quantize_intN_per_row(t, clip_range=clip) + result[name + ".q"] = q + result[name + ".scale"] = s + meta[name] = {"type": f"int{5 if cat == 'mlp' else 6}"} + else: + q, s = quantize_float_tensor(t) + result[name + ".q"] = q + result[name + ".scale"] = s + meta[name] = {"type": "int8"} + return result, meta + +def dequantize_mixed_int6(result: dict[str, Tensor], meta: dict[str, object], + template_sd: dict[str, Tensor]) -> dict[str, Tensor]: + out: dict[str, Tensor] = {} + for name, orig in template_sd.items(): + info = meta[name] + orig_dtype = orig.dtype + if info in ("passthrough", "passthrough_ctrl", "passthrough_fp16"): + t = result[name] + if t.dtype == torch.float16 and orig_dtype in (torch.float32, torch.bfloat16): + t = t.to(orig_dtype) + out[name] = t + continue + q, s = result[name + ".q"], result[name + ".scale"] + if s.ndim > 0: + out[name] = (q.float() * s.float().view(q.shape[0], *([1] * (q.ndim - 1)))).to(orig_dtype) + else: + out[name] = (q.float() * float(s.item())).to(orig_dtype) + return out + + +# ----------------------------- +# DATA LOADING +# ----------------------------- + +def load_data_shard(file: Path) -> Tensor: + header_bytes = 256 * np.dtype(" None: + self.file_idx = (self.file_idx + 1) % len(self.files) + self.tokens = load_data_shard(self.files[self.file_idx]) + self.pos = 0 + + def take(self, n: int) -> Tensor: + chunks: list[Tensor] = [] + remaining = n + while remaining > 0: + avail = self.tokens.numel() - self.pos + if avail <= 0: + self._advance_file() + continue + k = min(remaining, avail) + chunks.append(self.tokens[self.pos : self.pos + k]) + self.pos += k + remaining -= k + return chunks[0] if len(chunks) == 1 else torch.cat(chunks) + + +class DistributedTokenLoader: + def __init__(self, pattern: str, rank: int, world_size: int, device: torch.device): + self.rank = rank + self.world_size = world_size + self.device = device + self.stream = TokenStream(pattern) + + def next_batch(self, global_tokens: int, seq_len: int, grad_accum_steps: int) -> tuple[Tensor, Tensor]: + local_tokens = global_tokens // (self.world_size * grad_accum_steps) + per_rank_span = local_tokens + 1 + chunk = self.stream.take(per_rank_span * self.world_size) + start = self.rank * per_rank_span + local = chunk[start : start + per_rank_span].to(dtype=torch.int64) + x = local[:-1].reshape(-1, seq_len) + y = local[1:].reshape(-1, seq_len) + return x.to(self.device, non_blocking=True), y.to(self.device, non_blocking=True) + + +# ----------------------------- +# TRANSFORMER MODULES +# ----------------------------- + +class RMSNorm(nn.Module): + def __init__(self, eps: float | None = None): + super().__init__() + self.eps = eps + + def forward(self, x: Tensor) -> Tensor: + return F.rms_norm(x, (x.size(-1),), eps=self.eps) + + +class CastedLinear(nn.Linear): + def forward(self, x: Tensor) -> Tensor: + w = self.weight.to(x.dtype) + bias = self.bias.to(x.dtype) if self.bias is not None else None + return F.linear(x, w, bias) + + +def restore_low_dim_params_to_fp32(module: nn.Module) -> None: + with torch.no_grad(): + for name, param in module.named_parameters(): + if (param.ndim < 2 or any(pattern in name for pattern in CONTROL_TENSOR_NAME_PATTERNS)) and param.dtype != torch.float32: + param.data = param.data.float() + + +class Rotary(nn.Module): + def __init__(self, dim: int, base: float = 10000.0): + super().__init__() + inv_freq = 1.0 / (base ** (torch.arange(0, dim, 2, dtype=torch.float32) / dim)) + self.register_buffer("inv_freq", inv_freq, persistent=False) + self._seq_len_cached = 0 + self._cos_cached: Tensor | None = None + self._sin_cached: Tensor | None = None + + def forward(self, seq_len: int, device: torch.device, dtype: torch.dtype) -> tuple[Tensor, Tensor]: + if ( + self._cos_cached is None + or self._sin_cached is None + or self._seq_len_cached != seq_len + or self._cos_cached.device != device + ): + t = torch.arange(seq_len, device=device, dtype=self.inv_freq.dtype) + freqs = torch.outer(t, self.inv_freq.to(device)) + self._cos_cached = freqs.cos()[None, None, :, :] + self._sin_cached = freqs.sin()[None, None, :, :] + self._seq_len_cached = seq_len + return self._cos_cached.to(dtype=dtype), self._sin_cached.to(dtype=dtype) + + +def apply_rotary_emb(x: Tensor, cos: Tensor, sin: Tensor) -> Tensor: + half = x.size(-1) // 2 + x1, x2 = x[..., :half], x[..., half:] + return torch.cat((x1 * cos + x2 * sin, x1 * (-sin) + x2 * cos), dim=-1) + + +class CausalSelfAttention(nn.Module): + def __init__(self, dim: int, num_heads: int, num_kv_heads: int, rope_base: float, qk_gain_init: float, + rope_dims: int = 0, use_xsa: bool = False): + super().__init__() + if dim % num_heads != 0: + raise ValueError("model_dim must be divisible by num_heads") + if num_heads % num_kv_heads != 0: + raise ValueError("num_heads must be divisible by num_kv_heads") + self.num_heads = num_heads + self.num_kv_heads = num_kv_heads + self.head_dim = dim // num_heads + if self.head_dim % 2 != 0: + raise ValueError("head_dim must be even for RoPE") + self.rope_dims = rope_dims if rope_dims > 0 else self.head_dim + self.use_xsa = use_xsa + kv_dim = self.num_kv_heads * self.head_dim + self.c_q = CastedLinear(dim, dim, bias=False) + self.c_k = CastedLinear(dim, kv_dim, bias=False) + self.c_v = CastedLinear(dim, kv_dim, bias=False) + self.proj = CastedLinear(dim, dim, bias=False) + self.proj._zero_init = True + self.q_gain = nn.Parameter(torch.full((num_heads,), qk_gain_init, dtype=torch.float32)) + self.rotary = Rotary(self.rope_dims, base=rope_base) + + def _xsa_efficient(self, y: Tensor, v: Tensor) -> Tensor: + """Exclusive self-attention: subtract self-value from attention output.""" + # y is post-attention [bsz, heads, seq, head_dim], v is [bsz, kv_heads, seq, head_dim] + if self.num_kv_heads != self.num_heads: + v = v.repeat_interleave(self.num_heads // self.num_kv_heads, dim=1) + return y - v / v.size(2) + + def forward(self, x: Tensor, v0: Tensor | None = None) -> tuple[Tensor, Tensor]: + bsz, seqlen, dim = x.shape + q = self.c_q(x).reshape(bsz, seqlen, self.num_heads, self.head_dim).transpose(1, 2) + k = self.c_k(x).reshape(bsz, seqlen, self.num_kv_heads, self.head_dim).transpose(1, 2) + v = self.c_v(x).reshape(bsz, seqlen, self.num_kv_heads, self.head_dim).transpose(1, 2) + # Value Residual: blend with layer-0 V + if v0 is not None: + v = 0.5 * (v + v0) + v_out = v + q = F.rms_norm(q, (q.size(-1),)) + k = F.rms_norm(k, (k.size(-1),)) + cos, sin = self.rotary(seqlen, x.device, q.dtype) + if self.rope_dims < self.head_dim: + # Partial RoPE: rotate only first rope_dims, pass rest through + q_rope, q_pass = q[..., :self.rope_dims], q[..., self.rope_dims:] + k_rope, k_pass = k[..., :self.rope_dims], k[..., self.rope_dims:] + q_rope = apply_rotary_emb(q_rope, cos, sin) + k_rope = apply_rotary_emb(k_rope, cos, sin) + q = torch.cat([q_rope, q_pass], dim=-1) + k = torch.cat([k_rope, k_pass], dim=-1) + else: + q = apply_rotary_emb(q, cos, sin) + k = apply_rotary_emb(k, cos, sin) + q = q * self.q_gain.to(dtype=q.dtype)[None, :, None, None] + if self.num_kv_heads != self.num_heads: + n_rep = self.num_heads // self.num_kv_heads + k = k.repeat_interleave(n_rep, dim=1) + v_sdpa = v.repeat_interleave(n_rep, dim=1) + else: + v_sdpa = v + y = F.scaled_dot_product_attention( + q, k, v_sdpa, attn_mask=None, is_causal=True, + ) + if self.use_xsa: + y = self._xsa_efficient(y, v) + y = y.transpose(1, 2).contiguous().reshape(bsz, seqlen, dim) + return self.proj(y), v_out + + +class MLP(nn.Module): + def __init__(self, dim: int, mlp_mult: float): + super().__init__() + hidden = int(mlp_mult * dim) + self.fc = CastedLinear(dim, hidden, bias=False) + self.proj = CastedLinear(hidden, dim, bias=False) + self.proj._zero_init = True + + def forward(self, x: Tensor) -> Tensor: + x = F.leaky_relu(self.fc(x), negative_slope=0.5) + return self.proj(x.square()) + + +class SmearGate(nn.Module): + """Blend each token's embedding with the previous token's embedding.""" + def __init__(self, dim: int): + super().__init__() + self.gate = nn.Parameter(torch.zeros(dim, dtype=torch.float32)) + + def forward(self, x: Tensor) -> Tensor: + g = torch.sigmoid(self.gate.to(dtype=x.dtype))[None, None, :] + x_prev = torch.cat([torch.zeros_like(x[:, :1]), x[:, :-1]], dim=1) + return (1 - g) * x + g * x_prev + + +class BigramHashEmbedding(nn.Module): + """Hash consecutive token pairs into a learned embedding table.""" + def __init__(self, bigram_vocab_size: int, bigram_dim: int, model_dim: int): + super().__init__() + self.bigram_vocab_size = bigram_vocab_size + self.embed = nn.Embedding(bigram_vocab_size, bigram_dim) + nn.init.zeros_(self.embed.weight) + self.proj = CastedLinear(bigram_dim, model_dim, bias=False) if bigram_dim != model_dim else None + if self.proj is not None: + nn.init.zeros_(self.proj.weight) + self.scale = nn.Parameter(torch.tensor(0.05, dtype=torch.float32)) + + def bigram_hash(self, tokens: Tensor) -> Tensor: + t = tokens.to(torch.int32) + mod = self.bigram_vocab_size - 1 + out = torch.empty_like(t) + out[..., 0] = mod + out[..., 1:] = torch.bitwise_xor(36313 * t[..., 1:], 27191 * t[..., :-1]) % mod + return out.long() + + def forward(self, token_ids: Tensor) -> Tensor: + h = self.embed(self.bigram_hash(token_ids)) + if self.proj is not None: + h = self.proj(h) + return h * self.scale.to(dtype=h.dtype) + + +class Block(nn.Module): + def __init__(self, dim: int, num_heads: int, num_kv_heads: int, mlp_mult: float, rope_base: float, + qk_gain_init: float, rope_dims: int = 0, use_xsa: bool = False, ln_scale_factor: float = 1.0): + super().__init__() + self.attn_norm = RMSNorm() + self.mlp_norm = RMSNorm() + self.attn = CausalSelfAttention(dim, num_heads, num_kv_heads, rope_base, qk_gain_init, + rope_dims=rope_dims, use_xsa=use_xsa) + self.mlp = MLP(dim, mlp_mult) + self.attn_scale = nn.Parameter(torch.ones(dim, dtype=torch.float32)) + self.mlp_scale = nn.Parameter(torch.ones(dim, dtype=torch.float32)) + self.resid_mix = nn.Parameter(torch.stack((torch.ones(dim), torch.zeros(dim))).float()) + self.ln_scale_factor = ln_scale_factor + + def forward(self, x: Tensor, x0: Tensor, v0: Tensor | None = None) -> tuple[Tensor, Tensor]: + mix = self.resid_mix.to(dtype=x.dtype) + x = mix[0][None, None, :] * x + mix[1][None, None, :] * x0 + s = self.ln_scale_factor + attn_out, v_out = self.attn(self.attn_norm(x) * s, v0=v0) + x = x + self.attn_scale.to(dtype=x.dtype)[None, None, :] * attn_out + x = x + self.mlp_scale.to(dtype=x.dtype)[None, None, :] * self.mlp(self.mlp_norm(x) * s) + return x, v_out + + +class GPT(nn.Module): + def __init__( + self, + vocab_size: int, + num_layers: int, + model_dim: int, + num_heads: int, + num_kv_heads: int, + mlp_mult: float, + tie_embeddings: bool, + tied_embed_init_std: float, + logit_softcap: float, + rope_base: float, + qk_gain_init: float, + bigram_vocab_size: int = 0, + bigram_dim: int = 128, + rope_dims: int = 0, + ln_scale: bool = False, + xsa_last_n: int = 0, + ): + super().__init__() + if logit_softcap <= 0.0: + raise ValueError(f"logit_softcap must be positive, got {logit_softcap}") + self.tie_embeddings = tie_embeddings + self.tied_embed_init_std = tied_embed_init_std + self.logit_softcap = logit_softcap + self.tok_emb = nn.Embedding(vocab_size, model_dim) + self.bigram = BigramHashEmbedding(bigram_vocab_size, bigram_dim, model_dim) if bigram_vocab_size > 0 else None + self.num_encoder_layers = num_layers // 2 + self.num_decoder_layers = num_layers - self.num_encoder_layers + self.num_skip_weights = min(self.num_encoder_layers, self.num_decoder_layers) + self.skip_weights = nn.Parameter(torch.ones(self.num_skip_weights, model_dim, dtype=torch.float32)) + self.smear = SmearGate(model_dim) + self.blocks = nn.ModuleList([ + Block(model_dim, num_heads, num_kv_heads, mlp_mult, rope_base, qk_gain_init, + rope_dims=rope_dims, + use_xsa=(i >= num_layers - xsa_last_n) if xsa_last_n > 0 else False, + ln_scale_factor=1.0 / math.sqrt(i + 1) if ln_scale else 1.0) + for i in range(num_layers) + ]) + self.final_norm = RMSNorm() + self.lm_head = None if tie_embeddings else CastedLinear(model_dim, vocab_size, bias=False) + if self.lm_head is not None: + self.lm_head._zero_init = True + self._init_weights() + + def _init_weights(self) -> None: + if self.tie_embeddings: + nn.init.normal_(self.tok_emb.weight, mean=0.0, std=self.tied_embed_init_std) + num_layers = len(self.blocks) + for name, module in self.named_modules(): + if isinstance(module, nn.Linear): + if getattr(module, "_zero_init", False): + nn.init.zeros_(module.weight) + elif module.weight.ndim == 2 and module.weight.shape[0] >= 64 and module.weight.shape[1] >= 64: + nn.init.orthogonal_(module.weight, gain=1.0) + if ".proj." in name or name.endswith(".proj"): + with torch.no_grad(): + module.weight.mul_(1.0 / math.sqrt(2 * num_layers)) + + def _forward_body(self, input_ids: Tensor) -> Tensor: + x = self.tok_emb(input_ids) + if self.bigram is not None: + x = x + self.bigram(input_ids) + x = F.rms_norm(x, (x.size(-1),)) + x = self.smear(x) + x0 = x + v0: Tensor | None = None + skips: list[Tensor] = [] + for i in range(self.num_encoder_layers): + x, v_out = self.blocks[i](x, x0, v0=v0) + if v0 is None: + v0 = v_out + skips.append(x) + for i in range(self.num_decoder_layers): + if skips: + x = x + self.skip_weights[i].to(dtype=x.dtype)[None, None, :] * skips.pop() + x, _ = self.blocks[self.num_encoder_layers + i](x, x0, v0=v0) + return x + + def forward(self, input_ids: Tensor, target_ids: Tensor) -> Tensor: + x = self._forward_body(input_ids) + x = self.final_norm(x).reshape(-1, x.size(-1)) + targets = target_ids.reshape(-1) + if self.tie_embeddings: + logits_proj = F.linear(x, self.tok_emb.weight) + else: + if self.lm_head is None: + raise RuntimeError("lm_head is required when tie_embeddings=False") + logits_proj = self.lm_head(x) + logits = self.logit_softcap * torch.tanh(logits_proj / self.logit_softcap) + return F.cross_entropy(logits.float(), targets, reduction="mean") + + def forward_logits(self, input_ids: Tensor) -> Tensor: + x = self._forward_body(input_ids) + x = self.final_norm(x) + if self.tie_embeddings: + logits_proj = F.linear(x, self.tok_emb.weight) + else: + logits_proj = self.lm_head(x) + return self.logit_softcap * torch.tanh(logits_proj / self.logit_softcap) + + +def eval_val_sliding( + args: Hyperparameters, + base_model: nn.Module, + rank: int, + world_size: int, + device: torch.device, + val_tokens: Tensor, + base_bytes_lut: Tensor, + has_leading_space_lut: Tensor, + is_boundary_token_lut: Tensor, + stride: int, + batch_seqs: int = 32, +) -> tuple[float, float]: + seq_len = args.train_seq_len + total_tokens = val_tokens.numel() - 1 + window_starts = [ws for ws in range(0, total_tokens, stride) + if min(ws + seq_len, total_tokens) - ws >= stride or ws == 0] + total_windows = len(window_starts) + my_s = (total_windows * rank) // world_size + my_e = (total_windows * (rank + 1)) // world_size + my_windows = window_starts[my_s:my_e] + + loss_sum = torch.zeros((), device=device, dtype=torch.float64) + token_count = torch.zeros((), device=device, dtype=torch.float64) + byte_count = torch.zeros((), device=device, dtype=torch.float64) + + eval_start = time.perf_counter() + eval_budget_s = 570.0 # 30s margin from 10-min eval budget + base_model.eval() + with torch.inference_mode(): + for bi in range(0, len(my_windows), batch_seqs): + eval_elapsed = time.perf_counter() - eval_start + if eval_elapsed > eval_budget_s: + if rank == 0: + print(f" FAILSAFE: eval time {eval_elapsed:.0f}s exceeds {eval_budget_s}s budget, returning partial results", flush=True) + break + batch_ws = my_windows[bi:bi + batch_seqs] + bsz = len(batch_ws) + x_batch = torch.zeros(bsz, seq_len, dtype=torch.int64, device=device) + y_batch = torch.zeros(bsz, seq_len, dtype=torch.int64, device=device) + wlens: list[int] = [] + for i, ws in enumerate(batch_ws): + end = min(ws + seq_len, total_tokens) + wlen = end - ws + wlens.append(wlen) + chunk = val_tokens[ws:end + 1].to(dtype=torch.int64, device=device) + x_batch[i, :wlen] = chunk[:-1] + y_batch[i, :wlen] = chunk[1:] + with torch.autocast(device_type="cuda", dtype=torch.bfloat16): + logits = base_model.forward_logits(x_batch) + nll = F.cross_entropy( + logits.reshape(-1, logits.size(-1)).float(), + y_batch.reshape(-1), + reduction="none", + ).reshape(bsz, seq_len) + for i, ws in enumerate(batch_ws): + wlen = wlens[i] + s = 0 if ws == 0 else max(wlen - stride, 0) + scored_nll = nll[i, s:wlen].to(torch.float64) + loss_sum += scored_nll.sum() + token_count += float(wlen - s) + tgt = y_batch[i, s:wlen] + prev = x_batch[i, s:wlen] + tb = base_bytes_lut[tgt].to(torch.float64) + tb += (has_leading_space_lut[tgt] & ~is_boundary_token_lut[prev]).to(torch.float64) + byte_count += tb.sum() + if rank == 0 and (bi // batch_seqs) % 50 == 0: + done = min(bi + batch_seqs, len(my_windows)) + pct = done / len(my_windows) * 100 + running_bpb = 0.0 + if token_count.item() > 0: + rl = (loss_sum / token_count).item() + running_bpb = rl / math.log(2.0) * (token_count.item() / byte_count.item()) + print(f" sliding_eval [{pct:5.1f}%] {done}/{len(my_windows)} windows running_bpb={running_bpb:.6f}", flush=True) + + if dist.is_available() and dist.is_initialized(): + dist.all_reduce(loss_sum, op=dist.ReduceOp.SUM) + dist.all_reduce(token_count, op=dist.ReduceOp.SUM) + dist.all_reduce(byte_count, op=dist.ReduceOp.SUM) + + val_loss = (loss_sum / token_count).item() + bits_per_token = val_loss / math.log(2.0) + tokens_per_byte = token_count.item() / byte_count.item() + base_model.train() + return val_loss, bits_per_token * tokens_per_byte + + +def eval_val_sliding_ngram( + args: Hyperparameters, + base_model: nn.Module, + rank: int, + world_size: int, + device: torch.device, + val_tokens: Tensor, + base_bytes_lut: Tensor, + has_leading_space_lut: Tensor, + is_boundary_token_lut: Tensor, + stride: int, + batch_seqs: int = 32, +) -> tuple[float, float]: + """Sliding eval with multi-order n-gram backoff + entropy-adaptive alpha (score-first, legal).""" + seq_len = args.train_seq_len + total_tokens = val_tokens.numel() - 1 + max_order = args.ngram_eval_max_order + min_order = args.ngram_eval_min_order + buckets = args.ngram_eval_buckets + min_count = args.ngram_eval_min_count + use_entropy = args.ngram_eval_entropy + ent_base = args.ngram_eval_ent_base + ent_range = args.ngram_eval_ent_range + ent_scale = args.ngram_eval_ent_scale + ent_thresh = args.ngram_eval_ent_thresh + base_alpha = args.ngram_eval_alpha + n_orders = max_order - min_order + 1 + + window_starts = [ws for ws in range(0, total_tokens, stride) + if min(ws + seq_len, total_tokens) - ws >= stride or ws == 0] + total_windows = len(window_starts) + my_s = (total_windows * rank) // world_size + my_e = (total_windows * (rank + 1)) // world_size + my_windows = window_starts[my_s:my_e] + + val_np = val_tokens.numpy() + ctx_tables = [np.zeros((buckets,), dtype=np.uint32) for _ in range(n_orders)] + full_tables = [np.zeros((buckets,), dtype=np.uint32) for _ in range(n_orders)] + mask = np.uint64(buckets - 1) + primes = np.array( + [np.uint64(36313), np.uint64(27191), np.uint64(51647), np.uint64(81929), + np.uint64(131071), np.uint64(175447), np.uint64(209591)], + dtype=np.uint64, + ) + + if rank == 0: + print(f"ngram_cache:enabled orders={min_order}-{max_order} backoff " + f"entropy={use_entropy} alpha={base_alpha} " + f"ent_base={ent_base} ent_range={ent_range} " + f"min_count={min_count} buckets={buckets}", flush=True) + + loss_sum = 0.0 + token_count = 0.0 + byte_count = 0.0 + + eval_start = time.perf_counter() + eval_budget_s = 570.0 + # Pre-allocate eval buffers (avoid per-batch allocation) + x_buf = torch.zeros(batch_seqs, seq_len, dtype=torch.int64, device=device) + y_buf = torch.zeros(batch_seqs, seq_len, dtype=torch.int64, device=device) + base_model.eval() + # Compile eval path for faster inference + compiled_logits = torch.compile(base_model.forward_logits, dynamic=False, fullgraph=True) + with torch.inference_mode(): + for bi in range(0, len(my_windows), batch_seqs): + eval_elapsed = time.perf_counter() - eval_start + if eval_elapsed > eval_budget_s: + if rank == 0: + print(f" FAILSAFE: ngram eval time {eval_elapsed:.0f}s exceeds budget", flush=True) + break + batch_ws = my_windows[bi:bi + batch_seqs] + bsz = len(batch_ws) + x_batch = x_buf[:bsz] + y_batch = y_buf[:bsz] + x_batch.zero_() + y_batch.zero_() + wlens: list[int] = [] + for i, ws in enumerate(batch_ws): + end = min(ws + seq_len, total_tokens) + wlen = end - ws + wlens.append(wlen) + chunk = val_tokens[ws:end + 1].to(dtype=torch.int64, device=device) + x_batch[i, :wlen] = chunk[:-1] + y_batch[i, :wlen] = chunk[1:] + + with torch.autocast(device_type="cuda", dtype=torch.bfloat16): + logits = compiled_logits(x_batch) + nll = F.cross_entropy( + logits.reshape(-1, logits.size(-1)).float(), + y_batch.reshape(-1), reduction="none", + ).reshape(bsz, seq_len) + + for i, ws in enumerate(batch_ws): + wlen = wlens[i] + s = 0 if ws == 0 else max(wlen - stride, 0) + seg_len = wlen - s + if seg_len <= 0: + continue + + seg_nll = nll[i, s:wlen].to(torch.float64).cpu().numpy() + seg_model_p = np.exp(-seg_nll) + n_seg = len(seg_nll) + global_j = np.arange(ws + s + 1, ws + wlen + 1, dtype=np.int64) + + # Entropy-adaptive alpha + if use_entropy: + with torch.no_grad(): + lp = F.log_softmax(logits[i, s:wlen].float(), dim=-1) + seg_ent = -(lp.exp() * lp).sum(dim=-1).cpu().numpy() + alpha_per_tok = ent_base + ent_range / ( + 1.0 + np.exp(-ent_scale * (seg_ent - ent_thresh))) + + # Precompute hashes for all orders + order_data = [] + for oi in range(n_orders): + ctx_w = min_order + oi - 1 + valid = global_j >= ctx_w + if not valid.any(): + order_data.append(None) + continue + v_idx = np.nonzero(valid)[0] + jv = global_j[v_idx] + ctx_hash = np.zeros(len(jv), dtype=np.uint64) + for k in range(ctx_w): + tok = val_np[jv - (ctx_w - k)].astype(np.uint64) + ctx_hash ^= tok * primes[k % len(primes)] + ctx_key = (ctx_hash & mask).astype(np.int64) + tgt_np = val_np[jv].astype(np.uint64) + full_key = ((ctx_hash ^ (tgt_np * primes[ctx_w % len(primes)])) & mask).astype(np.int64) + order_data.append((v_idx, ctx_key, full_key)) + + # Multi-order backoff: highest order first + best_p_ng = np.full(n_seg, -1.0) + for oi in range(n_orders - 1, -1, -1): + if order_data[oi] is None: + continue + v_idx, ctx_key, full_key = order_data[oi] + ctx_counts = ctx_tables[oi][ctx_key].astype(np.float64) + full_counts = full_tables[oi][full_key].astype(np.float64) + has_match = (ctx_counts >= float(min_count)) & (full_counts > 0) + needs_fill = has_match & (best_p_ng[v_idx] < 0) + if needs_fill.any(): + fill_idx = v_idx[needs_fill] + p = np.minimum(full_counts[needs_fill], ctx_counts[needs_fill]) / np.maximum(ctx_counts[needs_fill], 1.0) + best_p_ng[fill_idx] = np.clip(p, 0.0, 1.0) + + # Mix model probability with n-gram + has_match = best_p_ng >= 0 + if has_match.any(): + if use_entropy: + alpha = alpha_per_tok[has_match] + else: + alpha = base_alpha + seg_model_p[has_match] = (1.0 - alpha) * seg_model_p[has_match] + alpha * best_p_ng[has_match] + seg_nll = -np.log(np.clip(seg_model_p, 1e-12, 1.0)) + + # Score-first: update ALL order tables AFTER scoring + for oi in range(n_orders): + if order_data[oi] is None: + continue + v_idx, ctx_key, full_key = order_data[oi] + np.add.at(ctx_tables[oi], ctx_key, 1) + np.add.at(full_tables[oi], full_key, 1) + + loss_sum += float(seg_nll.sum()) + token_count += float(seg_len) + tgt = y_batch[i, s:wlen] + prev = x_batch[i, s:wlen] + tb = base_bytes_lut[tgt].to(torch.float64) + tb += (has_leading_space_lut[tgt] & ~is_boundary_token_lut[prev]).to(torch.float64) + byte_count += float(tb.sum().item()) + + if rank == 0 and (bi // batch_seqs) % 200 == 0 and bi > 0: + done = min(bi + batch_seqs, len(my_windows)) + pct = done / len(my_windows) * 100 + cur_bpb = (loss_sum / max(token_count, 1.0)) / math.log(2.0) * (token_count / max(byte_count, 1.0)) + elapsed = time.perf_counter() - eval_start + print(f" ngram_eval [{pct:5.1f}%] bpb={cur_bpb:.6f} t={elapsed:.0f}s", flush=True) + + _loss = torch.tensor(loss_sum, device=device, dtype=torch.float64) + _toks = torch.tensor(token_count, device=device, dtype=torch.float64) + _bytes = torch.tensor(byte_count, device=device, dtype=torch.float64) + if dist.is_available() and dist.is_initialized(): + dist.all_reduce(_loss, op=dist.ReduceOp.SUM) + dist.all_reduce(_toks, op=dist.ReduceOp.SUM) + dist.all_reduce(_bytes, op=dist.ReduceOp.SUM) + + val_loss = _loss.item() / max(_toks.item(), 1.0) + val_bpb = val_loss / math.log(2.0) * (_toks.item() / max(_bytes.item(), 1.0)) + # Coverage check: warn if eval was cut short + total_expected = sum(1 for ws in window_starts + if (min(ws + seq_len, total_tokens) - ws - (0 if ws == 0 else max(min(ws + seq_len, total_tokens) - ws - stride, 0))) > 0) + coverage = _toks.item() / max(total_expected * stride, 1.0) # approximate + elapsed = time.perf_counter() - eval_start + if rank == 0: + print(f" ngram_eval DONE: bpb={val_bpb:.6f} tokens={_toks.item():.0f} t={elapsed:.0f}s", flush=True) + if elapsed >= eval_budget_s - 10: + print(f" WARNING: eval used {elapsed:.0f}s of {eval_budget_s}s budget — results may be from partial coverage", flush=True) + base_model.train() + return val_loss, val_bpb + + +# ----------------------------- +# TRAINING +# ----------------------------- + +def main() -> None: + global zeropower_via_newtonschulz5 + + code = Path(__file__).read_text(encoding="utf-8") + args = Hyperparameters() + zeropower_via_newtonschulz5 = torch.compile(zeropower_via_newtonschulz5) + + distributed = "RANK" in os.environ and "WORLD_SIZE" in os.environ + rank = int(os.environ.get("RANK", "0")) + world_size = int(os.environ.get("WORLD_SIZE", "1")) + local_rank = int(os.environ.get("LOCAL_RANK", "0")) + if world_size <= 0: + raise ValueError(f"WORLD_SIZE must be positive, got {world_size}") + if 8 % world_size != 0: + raise ValueError(f"WORLD_SIZE={world_size} must divide 8 so grad_accum_steps stays integral") + grad_accum_steps = 8 // world_size + grad_scale = 1.0 / grad_accum_steps + if not torch.cuda.is_available(): + raise RuntimeError("CUDA is required") + device = torch.device("cuda", local_rank) + torch.cuda.set_device(device) + if distributed: + dist.init_process_group(backend="nccl", device_id=device) + dist.barrier() + master_process = rank == 0 + + torch.backends.cuda.matmul.allow_tf32 = True + torch.backends.cudnn.allow_tf32 = True + try: + from torch.backends.cuda import enable_cudnn_sdp, enable_flash_sdp, enable_math_sdp, enable_mem_efficient_sdp + enable_cudnn_sdp(False) + enable_flash_sdp(True) + enable_mem_efficient_sdp(False) + enable_math_sdp(False) + except ImportError: + pass + + logfile = None + if master_process: + os.makedirs("logs", exist_ok=True) + logfile = f"logs/{args.run_id}.txt" + print(logfile) + + def log0(msg: str, console: bool = True) -> None: + if not master_process: + return + if console: + print(msg) + if logfile is not None: + with open(logfile, "a", encoding="utf-8") as f: + print(msg, file=f) + + log0(code, console=False) + log0("=" * 100, console=False) + log0(f"Running Python {sys.version}", console=False) + log0(f"Running PyTorch {torch.__version__}", console=False) + log0( + subprocess.run(["nvidia-smi"], stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True, check=False).stdout, + console=False, + ) + log0("=" * 100, console=False) + + random.seed(args.seed) + np.random.seed(args.seed) + torch.manual_seed(args.seed) + torch.cuda.manual_seed_all(args.seed) + + if not args.tokenizer_path.endswith(".model"): + raise ValueError(f"Script only setup for SentencePiece .model file: {args.tokenizer_path}") + sp = spm.SentencePieceProcessor(model_file=args.tokenizer_path) + if int(sp.vocab_size()) != args.vocab_size: + raise ValueError( + f"VOCAB_SIZE={args.vocab_size} does not match tokenizer vocab_size={int(sp.vocab_size())}" + ) + dataset_dir = Path(args.data_path).resolve() + actual_train_files = len(list(dataset_dir.glob("fineweb_train_*.bin"))) + val_tokens = load_validation_tokens(args.val_files, args.train_seq_len) + base_bytes_lut, has_leading_space_lut, is_boundary_token_lut = build_sentencepiece_luts( + sp, args.vocab_size, device + ) + log0(f"val_bpb:enabled tokenizer_kind=sentencepiece tokenizer_path={args.tokenizer_path}") + log0(f"train_loader:dataset:{dataset_dir.name} train_shards:{actual_train_files}") + log0(f"val_loader:shards pattern={args.val_files} tokens:{val_tokens.numel() - 1}") + + # MODEL + OPTIMIZER SETUP + base_model = GPT( + vocab_size=args.vocab_size, + num_layers=args.num_layers, + model_dim=args.model_dim, + num_heads=args.num_heads, + num_kv_heads=args.num_kv_heads, + mlp_mult=args.mlp_mult, + tie_embeddings=args.tie_embeddings, + tied_embed_init_std=args.tied_embed_init_std, + logit_softcap=args.logit_softcap, + rope_base=args.rope_base, + qk_gain_init=args.qk_gain_init, + bigram_vocab_size=args.bigram_vocab_size, + bigram_dim=args.bigram_dim, + rope_dims=args.rope_dims, + ln_scale=args.ln_scale, + xsa_last_n=args.xsa_last_n, + ).to(device).bfloat16() + for module in base_model.modules(): + if isinstance(module, CastedLinear): + module.float() + restore_low_dim_params_to_fp32(base_model) + torch._dynamo.config.optimize_ddp = False + compiled_model = torch.compile(base_model, dynamic=False, fullgraph=True) + model: nn.Module = DDP(compiled_model, device_ids=[local_rank], broadcast_buffers=False) if distributed else compiled_model + + block_named_params = list(base_model.blocks.named_parameters()) + matrix_params = [ + p for name, p in block_named_params + if p.ndim == 2 and not any(pattern in name for pattern in CONTROL_TENSOR_NAME_PATTERNS) + ] + scalar_params = [ + p for name, p in block_named_params + if p.ndim < 2 or any(pattern in name for pattern in CONTROL_TENSOR_NAME_PATTERNS) + ] + if base_model.skip_weights.numel() > 0: + scalar_params.append(base_model.skip_weights) + scalar_params.append(base_model.smear.gate) + if base_model.bigram is not None: + scalar_params.append(base_model.bigram.scale) + + token_lr = args.tied_embed_lr if args.tie_embeddings else args.embed_lr + tok_params = [{"params": [base_model.tok_emb.weight], "lr": token_lr, "base_lr": token_lr}] + if base_model.bigram is not None: + tok_params.append({"params": [base_model.bigram.embed.weight], "lr": token_lr, "base_lr": token_lr}) + if base_model.bigram.proj is not None: + matrix_params.append(base_model.bigram.proj.weight) + + optimizer_tok = torch.optim.AdamW( + tok_params, + betas=(args.beta1, args.beta2), + eps=args.adam_eps, + weight_decay=args.weight_decay, + fused=True, + ) + optimizer_muon = Muon( + matrix_params, + lr=args.matrix_lr, + momentum=args.muon_momentum, + backend_steps=args.muon_backend_steps, + weight_decay=0.04, + ) + for group in optimizer_muon.param_groups: + group["base_lr"] = args.matrix_lr + optimizer_scalar = torch.optim.AdamW( + [{"params": scalar_params, "lr": args.scalar_lr, "base_lr": args.scalar_lr}], + betas=(args.beta1, args.beta2), + eps=args.adam_eps, + weight_decay=args.weight_decay, + fused=True, + ) + optimizers: list[torch.optim.Optimizer] = [optimizer_tok, optimizer_muon, optimizer_scalar] + if base_model.lm_head is not None: + optimizer_head = torch.optim.Adam( + [{"params": [base_model.lm_head.weight], "lr": args.head_lr, "base_lr": args.head_lr}], + betas=(args.beta1, args.beta2), + eps=args.adam_eps, + fused=True, + ) + optimizers.insert(1, optimizer_head) + + n_params = sum(p.numel() for p in base_model.parameters()) + log0(f"model_params:{n_params}") + log0(f"world_size:{world_size} grad_accum_steps:{grad_accum_steps}") + log0(f"attention_mode:gqa num_heads:{args.num_heads} num_kv_heads:{args.num_kv_heads}") + log0( + f"tie_embeddings:{args.tie_embeddings} embed_lr:{token_lr} " + f"matrix_lr:{args.matrix_lr} scalar_lr:{args.scalar_lr}" + ) + log0( + f"train_batch_tokens:{args.train_batch_tokens} train_seq_len:{args.train_seq_len} " + f"iterations:{args.iterations} warmup_steps:{args.warmup_steps} " + f"max_wallclock_seconds:{args.max_wallclock_seconds:.3f}" + ) + log0(f"seed:{args.seed}") + + # DATA LOADER & MODEL WARMUP + train_loader = DistributedTokenLoader(args.train_files, rank, world_size, device) + + def zero_grad_all() -> None: + for opt in optimizers: + opt.zero_grad(set_to_none=True) + + max_wallclock_ms = 1000.0 * args.max_wallclock_seconds if args.max_wallclock_seconds > 0 else None + + def lr_mul(step: int, elapsed_ms: float) -> float: + if args.warmdown_iters <= 0: + return 1.0 + if max_wallclock_ms is None: + warmdown_start = max(args.iterations - args.warmdown_iters, 0) + return max((args.iterations - step) / max(args.warmdown_iters, 1), 0.0) if warmdown_start <= step < args.iterations else 1.0 + step_ms = elapsed_ms / max(step, 1) + warmdown_ms = args.warmdown_iters * step_ms + remaining_ms = max(max_wallclock_ms - elapsed_ms, 0.0) + return remaining_ms / max(warmdown_ms, 1e-9) if remaining_ms <= warmdown_ms else 1.0 + + if args.warmup_steps > 0: + initial_model_state = {name: tensor.detach().cpu().clone() for name, tensor in base_model.state_dict().items()} + initial_optimizer_states = [copy.deepcopy(opt.state_dict()) for opt in optimizers] + model.train() + for warmup_step in range(args.warmup_steps): + zero_grad_all() + for micro_step in range(grad_accum_steps): + if distributed: + model.require_backward_grad_sync = micro_step == grad_accum_steps - 1 + x, y = train_loader.next_batch(args.train_batch_tokens, args.train_seq_len, grad_accum_steps) + with torch.autocast(device_type="cuda", dtype=torch.bfloat16, enabled=True): + warmup_loss = model(x, y) + (warmup_loss * grad_scale).backward() + for opt in optimizers: + opt.step() + zero_grad_all() + if args.warmup_steps <= 20 or (warmup_step + 1) % 10 == 0 or warmup_step + 1 == args.warmup_steps: + log0(f"warmup_step:{warmup_step + 1}/{args.warmup_steps}") + base_model.load_state_dict(initial_model_state, strict=True) + for opt, state in zip(optimizers, initial_optimizer_states, strict=True): + opt.load_state_dict(state) + zero_grad_all() + if distributed: + model.require_backward_grad_sync = True + train_loader = DistributedTokenLoader(args.train_files, rank, world_size, device) + + # EMA shadow model (kept on GPU to avoid PCIe bottleneck) + ema_state: dict[str, Tensor] | None = None + if args.ema_enabled: + ema_state = {name: t.detach().clone() for name, t in base_model.state_dict().items()} + + # MAIN TRAINING LOOP + training_time_ms = 0.0 + stop_after_step: int | None = None + swa_state: dict[str, Tensor] | None = None + swa_count = 0 + torch.cuda.synchronize() + t0 = time.perf_counter() + + step = 0 + while True: + last_step = step == args.iterations or (stop_after_step is not None and step >= stop_after_step) + + should_validate = last_step or (args.val_loss_every > 0 and step % args.val_loss_every == 0) + if should_validate: + torch.cuda.synchronize() + training_time_ms += 1000.0 * (time.perf_counter() - t0) + val_loss, val_bpb = eval_val( + args, model, rank, world_size, device, grad_accum_steps, + val_tokens, base_bytes_lut, has_leading_space_lut, is_boundary_token_lut, + ) + log0( + f"step:{step}/{args.iterations} val_loss:{val_loss:.4f} val_bpb:{val_bpb:.4f} " + f"train_time:{training_time_ms:.0f}ms step_avg:{training_time_ms / max(step, 1):.2f}ms" + ) + torch.cuda.synchronize() + t0 = time.perf_counter() + + if last_step: + if stop_after_step is not None and step < args.iterations: + log0( + f"stopping_early: wallclock_cap train_time:{training_time_ms:.0f}ms " + f"step:{step}/{args.iterations}" + ) + break + + elapsed_ms = training_time_ms + 1000.0 * (time.perf_counter() - t0) + scale = lr_mul(step, elapsed_ms) + zero_grad_all() + train_loss = torch.zeros((), device=device) + for micro_step in range(grad_accum_steps): + if distributed: + model.require_backward_grad_sync = micro_step == grad_accum_steps - 1 + x, y = train_loader.next_batch(args.train_batch_tokens, args.train_seq_len, grad_accum_steps) + with torch.autocast(device_type="cuda", dtype=torch.bfloat16, enabled=True): + loss = model(x, y) + train_loss += loss.detach() + (loss * grad_scale).backward() + train_loss /= grad_accum_steps + + frac = min(step / args.muon_momentum_warmup_steps, 1.0) if args.muon_momentum_warmup_steps > 0 else 1.0 + muon_momentum = (1 - frac) * args.muon_momentum_warmup_start + frac * args.muon_momentum + for group in optimizer_muon.param_groups: + group["momentum"] = muon_momentum + + for opt in optimizers: + for group in opt.param_groups: + group["lr"] = group["base_lr"] * scale + + if args.grad_clip_norm > 0: + torch.nn.utils.clip_grad_norm_(base_model.parameters(), args.grad_clip_norm) + for opt in optimizers: + opt.step() + zero_grad_all() + + step += 1 + + # EMA update every 10 steps (GPU-resident, amortize overhead) + if ema_state is not None and step % 10 == 0: + decay = args.ema_decay ** 10 # compensate for batched updates + with torch.no_grad(): + for name, param in base_model.state_dict().items(): + ema_state[name].lerp_(param.detach(), 1.0 - decay) + + approx_training_time_ms = training_time_ms + 1000.0 * (time.perf_counter() - t0) + + # SWA: collect checkpoints during warmdown + if args.swa_enabled and scale < args.swa_start_frac and step % args.swa_every == 0: + if swa_state is None: + swa_state = {name: t.detach().cpu().clone() for name, t in base_model.state_dict().items()} + swa_count = 1 + log0(f"swa:start step:{step}") + else: + for name, t in base_model.state_dict().items(): + swa_state[name] += t.detach().cpu() + swa_count += 1 + + should_log_train = ( + args.train_log_every > 0 + and (step <= 10 or step % args.train_log_every == 0 or stop_after_step is not None) + ) + if should_log_train: + log0( + f"step:{step}/{args.iterations} train_loss:{train_loss.item():.4f} " + f"train_time:{approx_training_time_ms:.0f}ms step_avg:{approx_training_time_ms / step:.2f}ms" + ) + + reached_cap = max_wallclock_ms is not None and approx_training_time_ms >= max_wallclock_ms + if distributed and max_wallclock_ms is not None: + reached_cap_tensor = torch.tensor(int(reached_cap), device=device) + dist.all_reduce(reached_cap_tensor, op=dist.ReduceOp.MAX) + reached_cap = bool(reached_cap_tensor.item()) + if stop_after_step is None and reached_cap: + stop_after_step = step + + log0( + f"peak memory allocated: {torch.cuda.max_memory_allocated() // 1024 // 1024} MiB " + f"reserved: {torch.cuda.max_memory_reserved() // 1024 // 1024} MiB" + ) + + # Apply SWA if collected + if args.swa_enabled and swa_state is not None and swa_count > 1: + log0(f"swa:applying averaged {swa_count} checkpoints") + current_state = base_model.state_dict() + avg_state = { + name: (tensor / swa_count).to(dtype=current_state[name].dtype) + for name, tensor in swa_state.items() + } + base_model.load_state_dict(avg_state, strict=True) + + # Apply EMA if enabled (overrides SWA) + if args.ema_enabled and ema_state is not None: + log0("ema:applying shadow model") + current_state = base_model.state_dict() + ema_applied = { + name: tensor.to(dtype=current_state[name].dtype, device=current_state[name].device) + for name, tensor in ema_state.items() + } + base_model.load_state_dict(ema_applied, strict=True) + + # SERIALIZATION + ROUNDTRIP VALIDATION + if master_process: + torch.save(base_model.state_dict(), "final_model.pt") + model_bytes = os.path.getsize("final_model.pt") + code_bytes = len(code.encode("utf-8")) + log0(f"Serialized model: {model_bytes} bytes") + log0(f"Code size: {code_bytes} bytes") + log0(f"Total submission size: {model_bytes + code_bytes} bytes") + + # Magnitude pruning: zero out smallest weights to improve compression + with torch.no_grad(): + for name, param in base_model.named_parameters(): + if param.ndim == 2 and param.numel() > 65536: + threshold = torch.quantile(param.abs().float().flatten(), 0.03) + mask = param.abs() < threshold + param.masked_fill_(mask, 0.0) + + # INT6 mixed quantization + zstd/zlib export + sd_cpu = {k: v.detach().cpu() for k, v in base_model.state_dict().items()} + quant_result, quant_meta = mixed_quantize_int6(sd_cpu, {"mlp", "attn", "bigram"}) + quant_buf = io.BytesIO() + torch.save({"w": quant_result, "m": quant_meta}, quant_buf) + quant_raw = quant_buf.getvalue() + if _COMPRESSOR == "zstd": + quant_blob = zstandard.ZstdCompressor(level=22).compress(quant_raw) + else: + quant_blob = zlib.compress(quant_raw, 9) + if master_process: + with open("final_model.int8.ptz", "wb") as f: + f.write(quant_blob) + quant_file_bytes = os.path.getsize("final_model.int8.ptz") + code_bytes = len(code.encode("utf-8")) + log0(f"Serialized model int6+{_COMPRESSOR}: {quant_file_bytes} bytes") + total_bytes = quant_file_bytes + code_bytes + log0(f"Total submission size: {total_bytes} bytes ({total_bytes/1e6:.2f} MB)") + if total_bytes > 16_000_000: + log0(f"FAILSAFE: artifact {total_bytes} bytes EXCEEDS 16MB limit! Aborting eval.") + sys.exit(1) + log0(f"SIZE CHECK PASSED: {total_bytes/1e6:.2f} MB < 16.00 MB") + + if distributed: + dist.barrier() + with open("final_model.int8.ptz", "rb") as f: + quant_blob_disk = f.read() + if _COMPRESSOR == "zstd": + decompressed = zstandard.ZstdDecompressor().decompress(quant_blob_disk) + else: + decompressed = zlib.decompress(quant_blob_disk) + quant_state = torch.load(io.BytesIO(decompressed), map_location="cpu") + deq_state = dequantize_mixed_int6(quant_state["w"], quant_state["m"], sd_cpu) + base_model.load_state_dict(deq_state, strict=True) + + # Sliding window eval on int6-roundtripped weights + torch.cuda.synchronize() + t_qeval = time.perf_counter() + if args.ngram_eval_max_order >= 2 and args.eval_stride > 0: + log0(f"final_eval_mode:sliding_ngram orders={args.ngram_eval_min_order}-{args.ngram_eval_max_order} " + f"alpha={args.ngram_eval_alpha} entropy={args.ngram_eval_entropy} stride:{args.eval_stride}") + q_val_loss, q_val_bpb = eval_val_sliding_ngram( + args, base_model, rank, world_size, device, + val_tokens, base_bytes_lut, has_leading_space_lut, is_boundary_token_lut, + stride=args.eval_stride, + batch_seqs=args.eval_batch_seqs, + ) + elif args.eval_stride > 0 and args.eval_stride < args.train_seq_len: + log0(f"final_eval_mode:sliding_window stride:{args.eval_stride} batch_seqs:{args.eval_batch_seqs}") + q_val_loss, q_val_bpb = eval_val_sliding( + args, base_model, rank, world_size, device, + val_tokens, base_bytes_lut, has_leading_space_lut, is_boundary_token_lut, + stride=args.eval_stride, batch_seqs=args.eval_batch_seqs, + ) + else: + log0("final_eval_mode:standard") + q_val_loss, q_val_bpb = eval_val( + args, model, rank, world_size, device, grad_accum_steps, + val_tokens, base_bytes_lut, has_leading_space_lut, is_boundary_token_lut, + ) + torch.cuda.synchronize() + log0( + f"final_int8_zlib_roundtrip val_loss:{q_val_loss:.4f} val_bpb:{q_val_bpb:.4f} " + f"eval_time:{1000.0 * (time.perf_counter() - t_qeval):.0f}ms" + ) + log0(f"final_int8_zlib_roundtrip_exact val_loss:{q_val_loss:.8f} val_bpb:{q_val_bpb:.8f}") + + if distributed: + dist.destroy_process_group() + + +if __name__ == "__main__": + main() +# fixes applied +# tuned diff --git a/train_gpt.py b/train_gpt.py index 651beb2b8..7721b5a33 100644 --- a/train_gpt.py +++ b/train_gpt.py @@ -19,6 +19,12 @@ import zlib from pathlib import Path +try: + import zstandard + _COMPRESSOR = "zstd" +except ImportError: + _COMPRESSOR = "zlib" + import numpy as np import sentencepiece as spm import torch @@ -30,72 +36,93 @@ # ----------------------------- # HYPERPARAMETERS # ----------------------------- -# Default Simple Baseline run: -# - 9 transformer blocks at width 512 -# - 8 attention heads with 4 KV heads (GQA) and 2x MLP expansion -# - vocab size 1024, sequence length 1024, tied embeddings -# - 524,288 train tokens per step for 20,000 iterations with a ~10 minute cap class Hyperparameters: - # Data paths are shard globs produced by the existing preprocessing pipeline. data_path = os.environ.get("DATA_PATH", "./data/datasets/fineweb10B_sp1024") train_files = os.path.join(data_path, "fineweb_train_*.bin") val_files = os.path.join(data_path, "fineweb_val_*.bin") tokenizer_path = os.environ.get("TOKENIZER_PATH", "./data/tokenizers/fineweb_1024_bpe.model") run_id = os.environ.get("RUN_ID", str(uuid.uuid4())) - seed = int(os.environ.get("SEED", 1337)) + seed = int(os.environ.get("SEED", 42)) - # Validation cadence and batch size. Validation always uses the full fineweb_val split. val_batch_size = int(os.environ.get("VAL_BATCH_SIZE", 524_288)) - val_loss_every = int(os.environ.get("VAL_LOSS_EVERY", 1000)) - train_log_every = int(os.environ.get("TRAIN_LOG_EVERY", 200)) + val_loss_every = int(os.environ.get("VAL_LOSS_EVERY", 0)) # 0=skip mid-train val, maximize training steps + train_log_every = int(os.environ.get("TRAIN_LOG_EVERY", 100)) - # Training length. iterations = int(os.environ.get("ITERATIONS", 20000)) - warmdown_iters = int(os.environ.get("WARMDOWN_ITERS", 1200)) - warmup_steps = int(os.environ.get("WARMUP_STEPS", 20)) - train_batch_tokens = int(os.environ.get("TRAIN_BATCH_TOKENS", 524_288)) - train_seq_len = int(os.environ.get("TRAIN_SEQ_LEN", 1024)) + warmdown_iters = int(os.environ.get("WARMDOWN_ITERS", 3500)) + warmup_steps = int(os.environ.get("WARMUP_STEPS", 5)) # minimal warmup, maximize real steps + train_batch_tokens = int(os.environ.get("TRAIN_BATCH_TOKENS", 786_432)) + train_seq_len = int(os.environ.get("TRAIN_SEQ_LEN", 2048)) max_wallclock_seconds = float(os.environ.get("MAX_WALLCLOCK_SECONDS", 600.0)) qk_gain_init = float(os.environ.get("QK_GAIN_INIT", 1.5)) - # Model shape. vocab_size = int(os.environ.get("VOCAB_SIZE", 1024)) - num_layers = int(os.environ.get("NUM_LAYERS", 9)) + num_layers = int(os.environ.get("NUM_LAYERS", 10)) num_kv_heads = int(os.environ.get("NUM_KV_HEADS", 4)) model_dim = int(os.environ.get("MODEL_DIM", 512)) num_heads = int(os.environ.get("NUM_HEADS", 8)) - mlp_mult = int(os.environ.get("MLP_MULT", 2)) + mlp_mult = float(os.environ.get("MLP_MULT", 3.0)) tie_embeddings = bool(int(os.environ.get("TIE_EMBEDDINGS", "1"))) rope_base = float(os.environ.get("ROPE_BASE", 10000.0)) logit_softcap = float(os.environ.get("LOGIT_SOFTCAP", 30.0)) - # Optimizer hyperparameters. embed_lr = float(os.environ.get("EMBED_LR", 0.6)) head_lr = float(os.environ.get("HEAD_LR", 0.008)) - tied_embed_lr = float(os.environ.get("TIED_EMBED_LR", 0.05)) + tied_embed_lr = float(os.environ.get("TIED_EMBED_LR", 0.03)) tied_embed_init_std = float(os.environ.get("TIED_EMBED_INIT_STD", 0.005)) - matrix_lr = float(os.environ.get("MATRIX_LR", 0.04)) - scalar_lr = float(os.environ.get("SCALAR_LR", 0.04)) - muon_momentum = float(os.environ.get("MUON_MOMENTUM", 0.95)) + matrix_lr = float(os.environ.get("MATRIX_LR", 0.02)) + scalar_lr = float(os.environ.get("SCALAR_LR", 0.02)) + muon_momentum = float(os.environ.get("MUON_MOMENTUM", 0.99)) muon_backend_steps = int(os.environ.get("MUON_BACKEND_STEPS", 5)) - muon_momentum_warmup_start = float(os.environ.get("MUON_MOMENTUM_WARMUP_START", 0.85)) - muon_momentum_warmup_steps = int(os.environ.get("MUON_MOMENTUM_WARMUP_STEPS", 500)) + muon_momentum_warmup_start = float(os.environ.get("MUON_MOMENTUM_WARMUP_START", 0.92)) + muon_momentum_warmup_steps = int(os.environ.get("MUON_MOMENTUM_WARMUP_STEPS", 1500)) beta1 = float(os.environ.get("BETA1", 0.9)) beta2 = float(os.environ.get("BETA2", 0.95)) adam_eps = float(os.environ.get("ADAM_EPS", 1e-8)) - grad_clip_norm = float(os.environ.get("GRAD_CLIP_NORM", 0.0)) + grad_clip_norm = float(os.environ.get("GRAD_CLIP_NORM", 0.3)) + weight_decay = float(os.environ.get("WEIGHT_DECAY", 0.04)) + + eval_stride = int(os.environ.get("EVAL_STRIDE", 64)) + eval_batch_seqs = int(os.environ.get("EVAL_BATCH_SEQS", 64)) # larger batch for faster eval (no gradients) + + bigram_vocab_size = int(os.environ.get("BIGRAM_VOCAB_SIZE", 4096)) + bigram_dim = int(os.environ.get("BIGRAM_DIM", 128)) + + # Partial RoPE: only rotate first rope_dims dims (0 = full head_dim) + rope_dims = int(os.environ.get("ROPE_DIMS", 16)) + + # LN Scale: dampen norm inputs by 1/sqrt(layer_idx+1) for deeper layers + ln_scale = bool(int(os.environ.get("LN_SCALE", "1"))) + + # XSA: exclusive self-attention on last N layers (0 = disabled) + xsa_last_n = int(os.environ.get("XSA_LAST_N", 4)) # proven: last 4 layers + + # EMA: exponential moving average (replaces SWA when enabled) + ema_enabled = bool(int(os.environ.get("EMA_ENABLED", "1"))) + ema_decay = float(os.environ.get("EMA_DECAY", 0.997)) + + swa_enabled = bool(int(os.environ.get("SWA_ENABLED", "0"))) # OFF by default, EMA replaces it + swa_start_frac = float(os.environ.get("SWA_START_FRAC", 0.4)) + swa_every = int(os.environ.get("SWA_EVERY", 50)) + + # N-gram eval cache: multi-order backoff + entropy-adaptive alpha (score-first, legal) + ngram_eval_max_order = int(os.environ.get("NGRAM_EVAL_ORDER", 7)) # max n-gram order + ngram_eval_min_order = int(os.environ.get("NGRAM_EVAL_MIN_ORDER", 2)) # min backoff order + ngram_eval_alpha = float(os.environ.get("NGRAM_EVAL_ALPHA", 0.40)) # base alpha + ngram_eval_min_count = int(os.environ.get("NGRAM_EVAL_MIN_COUNT", 2)) + ngram_eval_buckets = int(os.environ.get("NGRAM_EVAL_BUCKETS", 4_194_304)) + ngram_eval_entropy = bool(int(os.environ.get("NGRAM_EVAL_ENTROPY", "1"))) + ngram_eval_ent_base = float(os.environ.get("NGRAM_EVAL_ENT_BASE", 0.05)) + ngram_eval_ent_range = float(os.environ.get("NGRAM_EVAL_ENT_RANGE", 0.55)) + ngram_eval_ent_scale = float(os.environ.get("NGRAM_EVAL_ENT_SCALE", 2.0)) + ngram_eval_ent_thresh = float(os.environ.get("NGRAM_EVAL_ENT_THRESH", 4.0)) # ----------------------------- -# MUON OPTIMIZER +# MUON OPTIMIZER # ----------------------------- -# -# As borrowed from modded-nanogpt -# Background on Muon: https://kellerjordan.github.io/posts/muon/ def zeropower_via_newtonschulz5(G: Tensor, steps: int = 10, eps: float = 1e-7) -> Tensor: - # Orthogonalize a 2D update matrix with a fast Newton-Schulz iteration. - # Muon uses this to normalize matrix-shaped gradients before applying them. a, b, c = (3.4445, -4.7750, 2.0315) X = G.bfloat16() X /= X.norm() + eps @@ -110,10 +137,10 @@ def zeropower_via_newtonschulz5(G: Tensor, steps: int = 10, eps: float = 1e-7) - class Muon(torch.optim.Optimizer): - def __init__(self, params, lr: float, momentum: float, backend_steps: int, nesterov: bool = True): + def __init__(self, params, lr: float, momentum: float, backend_steps: int, nesterov: bool = True, weight_decay: float = 0.0): super().__init__( params, - dict(lr=lr, momentum=momentum, backend_steps=backend_steps, nesterov=nesterov), + dict(lr=lr, momentum=momentum, backend_steps=backend_steps, nesterov=nesterov, weight_decay=weight_decay), ) @torch.no_grad() @@ -122,7 +149,6 @@ def step(self, closure=None): if closure is not None: with torch.enable_grad(): loss = closure() - distributed = dist.is_available() and dist.is_initialized() world_size = dist.get_world_size() if distributed else 1 rank = dist.get_rank() if distributed else 0 @@ -151,7 +177,6 @@ def step(self, closure=None): if nesterov: g = g.add(buf, alpha=momentum) g = zeropower_via_newtonschulz5(g, steps=backend_steps) - # Scale correction from Muon reference implementations. g *= max(1, g.size(0) / g.size(1)) ** 0.5 updates_flat[curr : curr + p.numel()] = g.reshape(-1) curr += p.numel() @@ -159,23 +184,20 @@ def step(self, closure=None): if distributed: dist.all_reduce(updates_flat, op=dist.ReduceOp.SUM) + wd = group.get("weight_decay", 0.0) curr = 0 for p in params: g = updates_flat[curr : curr + p.numel()].view_as(p).to(dtype=p.dtype) + if wd > 0: + p.data.mul_(1.0 - lr * wd) p.add_(g, alpha=-lr) curr += p.numel() - return loss # ----------------------------- -# TOKENIZER-AGNOSTIC EVALUATION SETUP +# TOKENIZER-AGNOSTIC EVALUATION # ----------------------------- -# -# It's common for small models have a large fraction of their parameters be embeddings, since the 2 * d_model * d_vocab vectors can be gigantic. -# Instead of locking the tokenizer, we let you bring your own and calculate our validation metrics on the average compression of the validation set. -# We calculate BPB (bits-per-byte) instead of validation loss, so we need methods to count the number of bits per token in the tokenizer. -# Note: Submissions that edit the tokenizer will be examined more carefully, since screwing this up might unjustly improve your score. def build_sentencepiece_luts( sp: spm.SentencePieceProcessor, vocab_size: int, device: torch.device @@ -193,7 +215,7 @@ def build_sentencepiece_luts( base_bytes_np[token_id] = 1 continue piece = sp.id_to_piece(token_id) - if piece.startswith("▁"): + if piece.startswith("\u2581"): has_leading_space_np[token_id] = True piece = piece[1:] base_bytes_np[token_id] = len(piece.encode("utf-8")) @@ -208,7 +230,6 @@ def load_validation_tokens(pattern: str, seq_len: int) -> Tensor: files = [Path(p) for p in sorted(glob.glob(pattern))] if not files: raise FileNotFoundError(f"No files found for pattern: {pattern}") - # The export pipeline writes the fixed first-50k-doc validation set to fineweb_val_*. tokens = torch.cat([load_data_shard(file) for file in files]).contiguous() usable = ((tokens.numel() - 1) // seq_len) * seq_len if usable <= 0: @@ -228,9 +249,6 @@ def eval_val( has_leading_space_lut: Tensor, is_boundary_token_lut: Tensor, ) -> tuple[float, float]: - # Validation computes two metrics: - # - val_loss: token cross-entropy (natural log) - # - val_bpb: tokenizer-agnostic compression metric used by the challenge local_batch_tokens = args.val_batch_size // (world_size * grad_accum_steps) if local_batch_tokens < args.train_seq_len: raise ValueError( @@ -245,7 +263,6 @@ def eval_val( val_loss_sum = torch.zeros((), device=device, dtype=torch.float64) val_token_count = torch.zeros((), device=device, dtype=torch.float64) val_byte_count = torch.zeros((), device=device, dtype=torch.float64) - model.eval() with torch.inference_mode(): for batch_seq_start in range(seq_start, seq_end, local_batch_seqs): @@ -265,34 +282,34 @@ def eval_val( token_bytes = base_bytes_lut[tgt_ids].to(dtype=torch.int16) token_bytes += (has_leading_space_lut[tgt_ids] & ~is_boundary_token_lut[prev_ids]).to(dtype=torch.int16) val_byte_count += token_bytes.to(torch.float64).sum() - if dist.is_available() and dist.is_initialized(): dist.all_reduce(val_loss_sum, op=dist.ReduceOp.SUM) dist.all_reduce(val_token_count, op=dist.ReduceOp.SUM) dist.all_reduce(val_byte_count, op=dist.ReduceOp.SUM) - val_loss = val_loss_sum / val_token_count bits_per_token = val_loss.item() / math.log(2.0) tokens_per_byte = val_token_count.item() / val_byte_count.item() model.train() return float(val_loss.item()), float(bits_per_token * tokens_per_byte) + # ----------------------------- -# POST-TRAINING QUANTIZATION +# POST-TRAINING QUANTIZATION (INT8 legacy + INT6 mixed) # ----------------------------- -# -# It's silly to export our model, which is trained in bf16 and fp32, at that same precision. -# Instead, we get approximately the same model (with a small hit) by quantizing the model to int8 & zlib compressing. -# We can then decompress the model and run in higher precision for evaluation, after closing in under the size limit. CONTROL_TENSOR_NAME_PATTERNS = tuple( pattern for pattern in os.environ.get( "CONTROL_TENSOR_NAME_PATTERNS", - "attn_scale,attn_scales,mlp_scale,mlp_scales,resid_mix,resid_mixes,q_gain,skip_weight,skip_weights", + "attn_scale,attn_scales,mlp_scale,mlp_scales,resid_mix,resid_mixes,q_gain,skip_weight,skip_weights,smear,bigram.scale", ).split(",") if pattern ) +FP16_KEEP_NAME_PATTERNS = tuple( + pattern + for pattern in os.environ.get("FP16_KEEP_NAME_PATTERNS", "tok_emb,blocks.8.attn.c_k").split(",") + if pattern +) INT8_KEEP_FLOAT_FP32_NAME_PATTERNS = tuple( pattern for pattern in os.environ.get( @@ -310,19 +327,9 @@ def eval_val( def tensor_nbytes(t: Tensor) -> int: return int(t.numel()) * int(t.element_size()) -def keep_float_tensor(name: str, t: Tensor, passthrough_orig_dtypes: dict[str, str]) -> Tensor: - if any(pattern in name for pattern in INT8_KEEP_FLOAT_FP32_NAME_PATTERNS): - return t.float().contiguous() - if t.dtype in {torch.float32, torch.bfloat16}: - passthrough_orig_dtypes[name] = str(t.dtype).removeprefix("torch.") - return t.to(dtype=INT8_KEEP_FLOAT_STORE_DTYPE).contiguous() - return t - def quantize_float_tensor(t: Tensor) -> tuple[Tensor, Tensor]: t32 = t.float() if t32.ndim == 2: - # Matrices get one scale per row, which usually tracks output-channel - # ranges much better than a single tensor-wide scale. clip_abs = ( torch.quantile(t32.abs(), INT8_CLIP_Q, dim=1) if t32.numel() @@ -332,105 +339,95 @@ def quantize_float_tensor(t: Tensor) -> tuple[Tensor, Tensor]: scale = (clip_abs / 127.0).clamp_min(1.0 / 127.0) q = torch.clamp(torch.round(clipped / scale[:, None]), -127, 127).to(torch.int8).contiguous() return q, scale.to(dtype=INT8_PER_ROW_SCALE_DTYPE).contiguous() - - # Vectors / scalars use a simpler per-tensor scale. clip_abs = float(torch.quantile(t32.abs().flatten(), INT8_CLIP_Q).item()) if t32.numel() else 0.0 scale = torch.tensor(clip_abs / 127.0 if clip_abs > 0 else 1.0, dtype=torch.float32) q = torch.clamp(torch.round(torch.clamp(t32, -clip_abs, clip_abs) / scale), -127, 127).to(torch.int8).contiguous() return q, scale -def quantize_state_dict_int8(state_dict: dict[str, Tensor]): - # Single supported clean-script export format: - # - per-row int8 for 2D float tensors - # - per-tensor int8 for other float tensors - # - exact passthrough for non-floats - # - passthrough for small float tensors, stored as fp16 to save bytes - quantized: dict[str, Tensor] = {} - scales: dict[str, Tensor] = {} - dtypes: dict[str, str] = {} - passthrough: dict[str, Tensor] = {} - passthrough_orig_dtypes: dict[str, str] = {} - qmeta: dict[str, dict[str, object]] = {} - stats = dict.fromkeys( - ("param_count", "num_tensors", "num_float_tensors", "num_nonfloat_tensors", "baseline_tensor_bytes", "int8_payload_bytes"), - 0, - ) +def _classify_param(name: str) -> str: + if "tok_emb" in name or "lm_head" in name: + return "embed" + if ".mlp." in name: + return "mlp" + if "bigram" in name: + return "bigram" + if ".attn." in name or (".proj." in name and ".mlp." not in name): + return "attn" + return "other" + +def quantize_intN_per_row(t: Tensor, clip_range: int = 31) -> tuple[Tensor, Tensor]: + t32 = t.float() + if t32.ndim == 2: + row_max = t32.abs().amax(dim=1) + scale = (row_max / clip_range).clamp_min(1e-12).to(torch.float16) + scale = scale.clamp_min(torch.finfo(torch.float16).tiny) + q = torch.clamp(torch.round(t32 / scale.float()[:, None]), -(clip_range+1), clip_range).to(torch.int8) + return q, scale + amax = t32.abs().max().item() + scale = torch.tensor(max(amax / clip_range, 1e-12), dtype=torch.float16) + q = torch.clamp(torch.round(t32 / scale.float()), -(clip_range+1), clip_range).to(torch.int8) + return q, scale + +def mixed_quantize_int6(state_dict: dict[str, Tensor], int6_cats: set[str]): + result: dict[str, Tensor] = {} + meta: dict[str, object] = {} for name, tensor in state_dict.items(): - t = tensor.detach().to("cpu").contiguous() - stats["param_count"] += int(t.numel()) - stats["num_tensors"] += 1 - stats["baseline_tensor_bytes"] += tensor_nbytes(t) - - if not t.is_floating_point(): - stats["num_nonfloat_tensors"] += 1 - passthrough[name] = t - stats["int8_payload_bytes"] += tensor_nbytes(t) + t = tensor.detach().cpu().contiguous() + cat = _classify_param(name) + if not t.is_floating_point() or t.numel() <= 8192: + result[name] = t.to(torch.float16) if t.is_floating_point() else t + meta[name] = "passthrough" continue - - # Small float tensors are cheap enough to keep directly. We still downcast - # fp32/bf16 passthrough tensors to fp16 so metadata does not dominate size. - if t.numel() <= INT8_KEEP_FLOAT_MAX_NUMEL: - kept = keep_float_tensor(name, t, passthrough_orig_dtypes) - passthrough[name] = kept - stats["int8_payload_bytes"] += tensor_nbytes(kept) + if any(p in name for p in CONTROL_TENSOR_NAME_PATTERNS): + result[name] = t.float() + meta[name] = "passthrough_ctrl" continue - - stats["num_float_tensors"] += 1 - q, s = quantize_float_tensor(t) - if s.ndim > 0: - qmeta[name] = {"scheme": "per_row", "axis": 0} - quantized[name] = q - scales[name] = s - dtypes[name] = str(t.dtype).removeprefix("torch.") - stats["int8_payload_bytes"] += tensor_nbytes(q) + tensor_nbytes(s) - - obj: dict[str, object] = { - "__quant_format__": "int8_clean_per_row_v1", - "quantized": quantized, - "scales": scales, - "dtypes": dtypes, - "passthrough": passthrough, - } - if qmeta: - obj["qmeta"] = qmeta - if passthrough_orig_dtypes: - obj["passthrough_orig_dtypes"] = passthrough_orig_dtypes - return obj, stats - -def dequantize_state_dict_int8(obj: dict[str, object]) -> dict[str, Tensor]: + if any(pattern in name for pattern in FP16_KEEP_NAME_PATTERNS): + result[name] = t.to(dtype=torch.float16).contiguous() + meta[name] = "passthrough_fp16" + continue + if cat in int6_cats and t.ndim >= 1: + clip = 15 if cat == "mlp" else 31 # int5 for MLP, int6 for attention + q, s = quantize_intN_per_row(t, clip_range=clip) + result[name + ".q"] = q + result[name + ".scale"] = s + meta[name] = {"type": f"int{5 if cat == 'mlp' else 6}"} + else: + q, s = quantize_float_tensor(t) + result[name + ".q"] = q + result[name + ".scale"] = s + meta[name] = {"type": "int8"} + return result, meta + +def dequantize_mixed_int6(result: dict[str, Tensor], meta: dict[str, object], + template_sd: dict[str, Tensor]) -> dict[str, Tensor]: out: dict[str, Tensor] = {} - qmeta = obj.get("qmeta", {}) - passthrough_orig_dtypes = obj.get("passthrough_orig_dtypes", {}) - for name, q in obj["quantized"].items(): - dtype = getattr(torch, obj["dtypes"][name]) - s = obj["scales"][name] - if qmeta.get(name, {}).get("scheme") == "per_row" or s.ndim > 0: - s = s.to(dtype=torch.float32) - # Broadcast the saved row scale back across trailing dimensions. - out[name] = (q.float() * s.view(q.shape[0], *([1] * (q.ndim - 1)))).to(dtype=dtype).contiguous() + for name, orig in template_sd.items(): + info = meta[name] + orig_dtype = orig.dtype + if info in ("passthrough", "passthrough_ctrl", "passthrough_fp16"): + t = result[name] + if t.dtype == torch.float16 and orig_dtype in (torch.float32, torch.bfloat16): + t = t.to(orig_dtype) + out[name] = t + continue + q, s = result[name + ".q"], result[name + ".scale"] + if s.ndim > 0: + out[name] = (q.float() * s.float().view(q.shape[0], *([1] * (q.ndim - 1)))).to(orig_dtype) else: - scale = float(s.item()) - out[name] = (q.float() * scale).to(dtype=dtype).contiguous() - for name, t in obj["passthrough"].items(): - # Restore small tensors, undoing the temporary fp16 storage cast if needed. - out_t = t.detach().to("cpu").contiguous() - orig_dtype = passthrough_orig_dtypes.get(name) - if isinstance(orig_dtype, str): - out_t = out_t.to(dtype=getattr(torch, orig_dtype)).contiguous() - out[name] = out_t + out[name] = (q.float() * float(s.item())).to(orig_dtype) return out # ----------------------------- -# DATA LOADING +# DATA LOADING # ----------------------------- def load_data_shard(file: Path) -> Tensor: header_bytes = 256 * np.dtype(" Tensor: class TokenStream: - # Reads shards sequentially and wraps around forever. The training loop therefore - # has deterministic, simple streaming behavior with no sampling or workers. def __init__(self, pattern: str): self.files = [Path(p) for p in sorted(glob.glob(pattern))] if not self.files: @@ -475,8 +470,6 @@ def take(self, n: int) -> Tensor: class DistributedTokenLoader: - # Each call consumes a contiguous chunk from the shared token stream, then slices out - # one disjoint span per rank. The extra "+1" token lets us build (x, y) by shifting. def __init__(self, pattern: str, rank: int, world_size: int, device: torch.device): self.rank = rank self.world_size = world_size @@ -493,6 +486,7 @@ def next_batch(self, global_tokens: int, seq_len: int, grad_accum_steps: int) -> y = local[1:].reshape(-1, seq_len) return x.to(self.device, non_blocking=True), y.to(self.device, non_blocking=True) + # ----------------------------- # TRANSFORMER MODULES # ----------------------------- @@ -507,14 +501,13 @@ def forward(self, x: Tensor) -> Tensor: class CastedLinear(nn.Linear): - # Keep weights in fp32 for optimizer/state quality, cast at matmul time for bf16 compute. def forward(self, x: Tensor) -> Tensor: + w = self.weight.to(x.dtype) bias = self.bias.to(x.dtype) if self.bias is not None else None - return F.linear(x, self.weight.to(x.dtype), bias) + return F.linear(x, w, bias) def restore_low_dim_params_to_fp32(module: nn.Module) -> None: - # Keep small/control parameters in fp32 even when the model body runs in bf16. with torch.no_grad(): for name, param in module.named_parameters(): if (param.ndim < 2 or any(pattern in name for pattern in CONTROL_TENSOR_NAME_PATTERNS)) and param.dtype != torch.float32: @@ -522,7 +515,6 @@ def restore_low_dim_params_to_fp32(module: nn.Module) -> None: class Rotary(nn.Module): - # Caches cos/sin tables per sequence length on the current device. def __init__(self, dim: int, base: float = 10000.0): super().__init__() inv_freq = 1.0 / (base ** (torch.arange(0, dim, 2, dtype=torch.float32) / dim)) @@ -553,14 +545,8 @@ def apply_rotary_emb(x: Tensor, cos: Tensor, sin: Tensor) -> Tensor: class CausalSelfAttention(nn.Module): - def __init__( - self, - dim: int, - num_heads: int, - num_kv_heads: int, - rope_base: float, - qk_gain_init: float, - ): + def __init__(self, dim: int, num_heads: int, num_kv_heads: int, rope_base: float, qk_gain_init: float, + rope_dims: int = 0, use_xsa: bool = False): super().__init__() if dim % num_heads != 0: raise ValueError("model_dim must be divisible by num_heads") @@ -571,6 +557,8 @@ def __init__( self.head_dim = dim // num_heads if self.head_dim % 2 != 0: raise ValueError("head_dim must be even for RoPE") + self.rope_dims = rope_dims if rope_dims > 0 else self.head_dim + self.use_xsa = use_xsa kv_dim = self.num_kv_heads * self.head_dim self.c_q = CastedLinear(dim, dim, bias=False) self.c_k = CastedLinear(dim, kv_dim, bias=False) @@ -578,71 +566,128 @@ def __init__( self.proj = CastedLinear(dim, dim, bias=False) self.proj._zero_init = True self.q_gain = nn.Parameter(torch.full((num_heads,), qk_gain_init, dtype=torch.float32)) - self.rotary = Rotary(self.head_dim, base=rope_base) + self.rotary = Rotary(self.rope_dims, base=rope_base) - def forward(self, x: Tensor) -> Tensor: + def _xsa_efficient(self, y: Tensor, v: Tensor) -> Tensor: + """Exclusive self-attention: subtract self-value from attention output.""" + # y is post-attention [bsz, heads, seq, head_dim], v is [bsz, kv_heads, seq, head_dim] + if self.num_kv_heads != self.num_heads: + v = v.repeat_interleave(self.num_heads // self.num_kv_heads, dim=1) + return y - v / v.size(2) + + def forward(self, x: Tensor, v0: Tensor | None = None) -> tuple[Tensor, Tensor]: bsz, seqlen, dim = x.shape q = self.c_q(x).reshape(bsz, seqlen, self.num_heads, self.head_dim).transpose(1, 2) k = self.c_k(x).reshape(bsz, seqlen, self.num_kv_heads, self.head_dim).transpose(1, 2) v = self.c_v(x).reshape(bsz, seqlen, self.num_kv_heads, self.head_dim).transpose(1, 2) + # Value Residual: blend with layer-0 V + if v0 is not None: + v = 0.5 * (v + v0) + v_out = v q = F.rms_norm(q, (q.size(-1),)) k = F.rms_norm(k, (k.size(-1),)) cos, sin = self.rotary(seqlen, x.device, q.dtype) - q = apply_rotary_emb(q, cos, sin) - k = apply_rotary_emb(k, cos, sin) + if self.rope_dims < self.head_dim: + # Partial RoPE: rotate only first rope_dims, pass rest through + q_rope, q_pass = q[..., :self.rope_dims], q[..., self.rope_dims:] + k_rope, k_pass = k[..., :self.rope_dims], k[..., self.rope_dims:] + q_rope = apply_rotary_emb(q_rope, cos, sin) + k_rope = apply_rotary_emb(k_rope, cos, sin) + q = torch.cat([q_rope, q_pass], dim=-1) + k = torch.cat([k_rope, k_pass], dim=-1) + else: + q = apply_rotary_emb(q, cos, sin) + k = apply_rotary_emb(k, cos, sin) q = q * self.q_gain.to(dtype=q.dtype)[None, :, None, None] + if self.num_kv_heads != self.num_heads: + n_rep = self.num_heads // self.num_kv_heads + k = k.repeat_interleave(n_rep, dim=1) + v_sdpa = v.repeat_interleave(n_rep, dim=1) + else: + v_sdpa = v y = F.scaled_dot_product_attention( - q, - k, - v, - attn_mask=None, - is_causal=True, - enable_gqa=(self.num_kv_heads != self.num_heads), + q, k, v_sdpa, attn_mask=None, is_causal=True, ) + if self.use_xsa: + y = self._xsa_efficient(y, v) y = y.transpose(1, 2).contiguous().reshape(bsz, seqlen, dim) - return self.proj(y) + return self.proj(y), v_out class MLP(nn.Module): - # relu^2 MLP from the original modded-nanogpt setup - def __init__(self, dim: int, mlp_mult: int): + def __init__(self, dim: int, mlp_mult: float): super().__init__() - hidden = mlp_mult * dim + hidden = int(mlp_mult * dim) self.fc = CastedLinear(dim, hidden, bias=False) self.proj = CastedLinear(hidden, dim, bias=False) self.proj._zero_init = True def forward(self, x: Tensor) -> Tensor: - x = torch.relu(self.fc(x)) + x = F.leaky_relu(self.fc(x), negative_slope=0.5) return self.proj(x.square()) +class SmearGate(nn.Module): + """Blend each token's embedding with the previous token's embedding.""" + def __init__(self, dim: int): + super().__init__() + self.gate = nn.Parameter(torch.zeros(dim, dtype=torch.float32)) + + def forward(self, x: Tensor) -> Tensor: + g = torch.sigmoid(self.gate.to(dtype=x.dtype))[None, None, :] + x_prev = torch.cat([torch.zeros_like(x[:, :1]), x[:, :-1]], dim=1) + return (1 - g) * x + g * x_prev + + +class BigramHashEmbedding(nn.Module): + """Hash consecutive token pairs into a learned embedding table.""" + def __init__(self, bigram_vocab_size: int, bigram_dim: int, model_dim: int): + super().__init__() + self.bigram_vocab_size = bigram_vocab_size + self.embed = nn.Embedding(bigram_vocab_size, bigram_dim) + nn.init.zeros_(self.embed.weight) + self.proj = CastedLinear(bigram_dim, model_dim, bias=False) if bigram_dim != model_dim else None + if self.proj is not None: + nn.init.zeros_(self.proj.weight) + self.scale = nn.Parameter(torch.tensor(0.05, dtype=torch.float32)) + + def bigram_hash(self, tokens: Tensor) -> Tensor: + t = tokens.to(torch.int32) + mod = self.bigram_vocab_size - 1 + out = torch.empty_like(t) + out[..., 0] = mod + out[..., 1:] = torch.bitwise_xor(36313 * t[..., 1:], 27191 * t[..., :-1]) % mod + return out.long() + + def forward(self, token_ids: Tensor) -> Tensor: + h = self.embed(self.bigram_hash(token_ids)) + if self.proj is not None: + h = self.proj(h) + return h * self.scale.to(dtype=h.dtype) + + class Block(nn.Module): - def __init__( - self, - dim: int, - num_heads: int, - num_kv_heads: int, - mlp_mult: int, - rope_base: float, - qk_gain_init: float, - ): + def __init__(self, dim: int, num_heads: int, num_kv_heads: int, mlp_mult: float, rope_base: float, + qk_gain_init: float, rope_dims: int = 0, use_xsa: bool = False, ln_scale_factor: float = 1.0): super().__init__() self.attn_norm = RMSNorm() self.mlp_norm = RMSNorm() - self.attn = CausalSelfAttention(dim, num_heads, num_kv_heads, rope_base, qk_gain_init) + self.attn = CausalSelfAttention(dim, num_heads, num_kv_heads, rope_base, qk_gain_init, + rope_dims=rope_dims, use_xsa=use_xsa) self.mlp = MLP(dim, mlp_mult) self.attn_scale = nn.Parameter(torch.ones(dim, dtype=torch.float32)) self.mlp_scale = nn.Parameter(torch.ones(dim, dtype=torch.float32)) self.resid_mix = nn.Parameter(torch.stack((torch.ones(dim), torch.zeros(dim))).float()) + self.ln_scale_factor = ln_scale_factor - def forward(self, x: Tensor, x0: Tensor) -> Tensor: + def forward(self, x: Tensor, x0: Tensor, v0: Tensor | None = None) -> tuple[Tensor, Tensor]: mix = self.resid_mix.to(dtype=x.dtype) x = mix[0][None, None, :] * x + mix[1][None, None, :] * x0 - attn_out = self.attn(self.attn_norm(x)) + s = self.ln_scale_factor + attn_out, v_out = self.attn(self.attn_norm(x) * s, v0=v0) x = x + self.attn_scale.to(dtype=x.dtype)[None, None, :] * attn_out - x = x + self.mlp_scale.to(dtype=x.dtype)[None, None, :] * self.mlp(self.mlp_norm(x)) - return x + x = x + self.mlp_scale.to(dtype=x.dtype)[None, None, :] * self.mlp(self.mlp_norm(x) * s) + return x, v_out class GPT(nn.Module): @@ -653,12 +698,17 @@ def __init__( model_dim: int, num_heads: int, num_kv_heads: int, - mlp_mult: int, + mlp_mult: float, tie_embeddings: bool, tied_embed_init_std: float, logit_softcap: float, rope_base: float, qk_gain_init: float, + bigram_vocab_size: int = 0, + bigram_dim: int = 128, + rope_dims: int = 0, + ln_scale: bool = False, + xsa_last_n: int = 0, ): super().__init__() if logit_softcap <= 0.0: @@ -667,23 +717,19 @@ def __init__( self.tied_embed_init_std = tied_embed_init_std self.logit_softcap = logit_softcap self.tok_emb = nn.Embedding(vocab_size, model_dim) + self.bigram = BigramHashEmbedding(bigram_vocab_size, bigram_dim, model_dim) if bigram_vocab_size > 0 else None self.num_encoder_layers = num_layers // 2 self.num_decoder_layers = num_layers - self.num_encoder_layers self.num_skip_weights = min(self.num_encoder_layers, self.num_decoder_layers) self.skip_weights = nn.Parameter(torch.ones(self.num_skip_weights, model_dim, dtype=torch.float32)) - self.blocks = nn.ModuleList( - [ - Block( - model_dim, - num_heads, - num_kv_heads, - mlp_mult, - rope_base, - qk_gain_init, - ) - for i in range(num_layers) - ] - ) + self.smear = SmearGate(model_dim) + self.blocks = nn.ModuleList([ + Block(model_dim, num_heads, num_kv_heads, mlp_mult, rope_base, qk_gain_init, + rope_dims=rope_dims, + use_xsa=(i >= num_layers - xsa_last_n) if xsa_last_n > 0 else False, + ln_scale_factor=1.0 / math.sqrt(i + 1) if ln_scale else 1.0) + for i in range(num_layers) + ]) self.final_norm = RMSNorm() self.lm_head = None if tie_embeddings else CastedLinear(model_dim, vocab_size, bias=False) if self.lm_head is not None: @@ -693,25 +739,39 @@ def __init__( def _init_weights(self) -> None: if self.tie_embeddings: nn.init.normal_(self.tok_emb.weight, mean=0.0, std=self.tied_embed_init_std) - for module in self.modules(): - if isinstance(module, nn.Linear) and getattr(module, "_zero_init", False): - nn.init.zeros_(module.weight) - - def forward(self, input_ids: Tensor, target_ids: Tensor) -> Tensor: + num_layers = len(self.blocks) + for name, module in self.named_modules(): + if isinstance(module, nn.Linear): + if getattr(module, "_zero_init", False): + nn.init.zeros_(module.weight) + elif module.weight.ndim == 2 and module.weight.shape[0] >= 64 and module.weight.shape[1] >= 64: + nn.init.orthogonal_(module.weight, gain=1.0) + if ".proj." in name or name.endswith(".proj"): + with torch.no_grad(): + module.weight.mul_(1.0 / math.sqrt(2 * num_layers)) + + def _forward_body(self, input_ids: Tensor) -> Tensor: x = self.tok_emb(input_ids) + if self.bigram is not None: + x = x + self.bigram(input_ids) x = F.rms_norm(x, (x.size(-1),)) + x = self.smear(x) x0 = x + v0: Tensor | None = None skips: list[Tensor] = [] - - # First half stores skips; second half reuses them in reverse order. for i in range(self.num_encoder_layers): - x = self.blocks[i](x, x0) + x, v_out = self.blocks[i](x, x0, v0=v0) + if v0 is None: + v0 = v_out skips.append(x) for i in range(self.num_decoder_layers): if skips: x = x + self.skip_weights[i].to(dtype=x.dtype)[None, None, :] * skips.pop() - x = self.blocks[self.num_encoder_layers + i](x, x0) + x, _ = self.blocks[self.num_encoder_layers + i](x, x0, v0=v0) + return x + def forward(self, input_ids: Tensor, target_ids: Tensor) -> Tensor: + x = self._forward_body(input_ids) x = self.final_norm(x).reshape(-1, x.size(-1)) targets = target_ids.reshape(-1) if self.tie_embeddings: @@ -723,6 +783,304 @@ def forward(self, input_ids: Tensor, target_ids: Tensor) -> Tensor: logits = self.logit_softcap * torch.tanh(logits_proj / self.logit_softcap) return F.cross_entropy(logits.float(), targets, reduction="mean") + def forward_logits(self, input_ids: Tensor) -> Tensor: + x = self._forward_body(input_ids) + x = self.final_norm(x) + if self.tie_embeddings: + logits_proj = F.linear(x, self.tok_emb.weight) + else: + logits_proj = self.lm_head(x) + return self.logit_softcap * torch.tanh(logits_proj / self.logit_softcap) + + +def eval_val_sliding( + args: Hyperparameters, + base_model: nn.Module, + rank: int, + world_size: int, + device: torch.device, + val_tokens: Tensor, + base_bytes_lut: Tensor, + has_leading_space_lut: Tensor, + is_boundary_token_lut: Tensor, + stride: int, + batch_seqs: int = 32, +) -> tuple[float, float]: + seq_len = args.train_seq_len + total_tokens = val_tokens.numel() - 1 + window_starts = [ws for ws in range(0, total_tokens, stride) + if min(ws + seq_len, total_tokens) - ws >= stride or ws == 0] + total_windows = len(window_starts) + my_s = (total_windows * rank) // world_size + my_e = (total_windows * (rank + 1)) // world_size + my_windows = window_starts[my_s:my_e] + + loss_sum = torch.zeros((), device=device, dtype=torch.float64) + token_count = torch.zeros((), device=device, dtype=torch.float64) + byte_count = torch.zeros((), device=device, dtype=torch.float64) + + eval_start = time.perf_counter() + eval_budget_s = 570.0 # 30s margin from 10-min eval budget + base_model.eval() + with torch.inference_mode(): + for bi in range(0, len(my_windows), batch_seqs): + eval_elapsed = time.perf_counter() - eval_start + if eval_elapsed > eval_budget_s: + if rank == 0: + print(f" FAILSAFE: eval time {eval_elapsed:.0f}s exceeds {eval_budget_s}s budget, returning partial results", flush=True) + break + batch_ws = my_windows[bi:bi + batch_seqs] + bsz = len(batch_ws) + x_batch = torch.zeros(bsz, seq_len, dtype=torch.int64, device=device) + y_batch = torch.zeros(bsz, seq_len, dtype=torch.int64, device=device) + wlens: list[int] = [] + for i, ws in enumerate(batch_ws): + end = min(ws + seq_len, total_tokens) + wlen = end - ws + wlens.append(wlen) + chunk = val_tokens[ws:end + 1].to(dtype=torch.int64, device=device) + x_batch[i, :wlen] = chunk[:-1] + y_batch[i, :wlen] = chunk[1:] + with torch.autocast(device_type="cuda", dtype=torch.bfloat16): + logits = base_model.forward_logits(x_batch) + nll = F.cross_entropy( + logits.reshape(-1, logits.size(-1)).float(), + y_batch.reshape(-1), + reduction="none", + ).reshape(bsz, seq_len) + for i, ws in enumerate(batch_ws): + wlen = wlens[i] + s = 0 if ws == 0 else max(wlen - stride, 0) + scored_nll = nll[i, s:wlen].to(torch.float64) + loss_sum += scored_nll.sum() + token_count += float(wlen - s) + tgt = y_batch[i, s:wlen] + prev = x_batch[i, s:wlen] + tb = base_bytes_lut[tgt].to(torch.float64) + tb += (has_leading_space_lut[tgt] & ~is_boundary_token_lut[prev]).to(torch.float64) + byte_count += tb.sum() + if rank == 0 and (bi // batch_seqs) % 50 == 0: + done = min(bi + batch_seqs, len(my_windows)) + pct = done / len(my_windows) * 100 + running_bpb = 0.0 + if token_count.item() > 0: + rl = (loss_sum / token_count).item() + running_bpb = rl / math.log(2.0) * (token_count.item() / byte_count.item()) + print(f" sliding_eval [{pct:5.1f}%] {done}/{len(my_windows)} windows running_bpb={running_bpb:.6f}", flush=True) + + if dist.is_available() and dist.is_initialized(): + dist.all_reduce(loss_sum, op=dist.ReduceOp.SUM) + dist.all_reduce(token_count, op=dist.ReduceOp.SUM) + dist.all_reduce(byte_count, op=dist.ReduceOp.SUM) + + val_loss = (loss_sum / token_count).item() + bits_per_token = val_loss / math.log(2.0) + tokens_per_byte = token_count.item() / byte_count.item() + base_model.train() + return val_loss, bits_per_token * tokens_per_byte + + +def eval_val_sliding_ngram( + args: Hyperparameters, + base_model: nn.Module, + rank: int, + world_size: int, + device: torch.device, + val_tokens: Tensor, + base_bytes_lut: Tensor, + has_leading_space_lut: Tensor, + is_boundary_token_lut: Tensor, + stride: int, + batch_seqs: int = 32, +) -> tuple[float, float]: + """Sliding eval with multi-order n-gram backoff + entropy-adaptive alpha (score-first, legal).""" + seq_len = args.train_seq_len + total_tokens = val_tokens.numel() - 1 + max_order = args.ngram_eval_max_order + min_order = args.ngram_eval_min_order + buckets = args.ngram_eval_buckets + min_count = args.ngram_eval_min_count + use_entropy = args.ngram_eval_entropy + ent_base = args.ngram_eval_ent_base + ent_range = args.ngram_eval_ent_range + ent_scale = args.ngram_eval_ent_scale + ent_thresh = args.ngram_eval_ent_thresh + base_alpha = args.ngram_eval_alpha + n_orders = max_order - min_order + 1 + + window_starts = [ws for ws in range(0, total_tokens, stride) + if min(ws + seq_len, total_tokens) - ws >= stride or ws == 0] + total_windows = len(window_starts) + my_s = (total_windows * rank) // world_size + my_e = (total_windows * (rank + 1)) // world_size + my_windows = window_starts[my_s:my_e] + + val_np = val_tokens.numpy() + ctx_tables = [np.zeros((buckets,), dtype=np.uint32) for _ in range(n_orders)] + full_tables = [np.zeros((buckets,), dtype=np.uint32) for _ in range(n_orders)] + mask = np.uint64(buckets - 1) + primes = np.array( + [np.uint64(36313), np.uint64(27191), np.uint64(51647), np.uint64(81929), + np.uint64(131071), np.uint64(175447), np.uint64(209591)], + dtype=np.uint64, + ) + + if rank == 0: + print(f"ngram_cache:enabled orders={min_order}-{max_order} backoff " + f"entropy={use_entropy} alpha={base_alpha} " + f"ent_base={ent_base} ent_range={ent_range} " + f"min_count={min_count} buckets={buckets}", flush=True) + + loss_sum = 0.0 + token_count = 0.0 + byte_count = 0.0 + + eval_start = time.perf_counter() + eval_budget_s = 570.0 + # Pre-allocate eval buffers (avoid per-batch allocation) + x_buf = torch.zeros(batch_seqs, seq_len, dtype=torch.int64, device=device) + y_buf = torch.zeros(batch_seqs, seq_len, dtype=torch.int64, device=device) + base_model.eval() + # Compile eval path for faster inference + compiled_logits = torch.compile(base_model.forward_logits, dynamic=False, fullgraph=True) + with torch.inference_mode(): + for bi in range(0, len(my_windows), batch_seqs): + eval_elapsed = time.perf_counter() - eval_start + if eval_elapsed > eval_budget_s: + if rank == 0: + print(f" FAILSAFE: ngram eval time {eval_elapsed:.0f}s exceeds budget", flush=True) + break + batch_ws = my_windows[bi:bi + batch_seqs] + bsz = len(batch_ws) + x_batch = x_buf[:bsz] + y_batch = y_buf[:bsz] + x_batch.zero_() + y_batch.zero_() + wlens: list[int] = [] + for i, ws in enumerate(batch_ws): + end = min(ws + seq_len, total_tokens) + wlen = end - ws + wlens.append(wlen) + chunk = val_tokens[ws:end + 1].to(dtype=torch.int64, device=device) + x_batch[i, :wlen] = chunk[:-1] + y_batch[i, :wlen] = chunk[1:] + + with torch.autocast(device_type="cuda", dtype=torch.bfloat16): + logits = compiled_logits(x_batch) + nll = F.cross_entropy( + logits.reshape(-1, logits.size(-1)).float(), + y_batch.reshape(-1), reduction="none", + ).reshape(bsz, seq_len) + + for i, ws in enumerate(batch_ws): + wlen = wlens[i] + s = 0 if ws == 0 else max(wlen - stride, 0) + seg_len = wlen - s + if seg_len <= 0: + continue + + seg_nll = nll[i, s:wlen].to(torch.float64).cpu().numpy() + seg_model_p = np.exp(-seg_nll) + n_seg = len(seg_nll) + global_j = np.arange(ws + s + 1, ws + wlen + 1, dtype=np.int64) + + # Entropy-adaptive alpha + if use_entropy: + with torch.no_grad(): + lp = F.log_softmax(logits[i, s:wlen].float(), dim=-1) + seg_ent = -(lp.exp() * lp).sum(dim=-1).cpu().numpy() + alpha_per_tok = ent_base + ent_range / ( + 1.0 + np.exp(-ent_scale * (seg_ent - ent_thresh))) + + # Precompute hashes for all orders + order_data = [] + for oi in range(n_orders): + ctx_w = min_order + oi - 1 + valid = global_j >= ctx_w + if not valid.any(): + order_data.append(None) + continue + v_idx = np.nonzero(valid)[0] + jv = global_j[v_idx] + ctx_hash = np.zeros(len(jv), dtype=np.uint64) + for k in range(ctx_w): + tok = val_np[jv - (ctx_w - k)].astype(np.uint64) + ctx_hash ^= tok * primes[k % len(primes)] + ctx_key = (ctx_hash & mask).astype(np.int64) + tgt_np = val_np[jv].astype(np.uint64) + full_key = ((ctx_hash ^ (tgt_np * primes[ctx_w % len(primes)])) & mask).astype(np.int64) + order_data.append((v_idx, ctx_key, full_key)) + + # Multi-order backoff: highest order first + best_p_ng = np.full(n_seg, -1.0) + for oi in range(n_orders - 1, -1, -1): + if order_data[oi] is None: + continue + v_idx, ctx_key, full_key = order_data[oi] + ctx_counts = ctx_tables[oi][ctx_key].astype(np.float64) + full_counts = full_tables[oi][full_key].astype(np.float64) + has_match = (ctx_counts >= float(min_count)) & (full_counts > 0) + needs_fill = has_match & (best_p_ng[v_idx] < 0) + if needs_fill.any(): + fill_idx = v_idx[needs_fill] + p = np.minimum(full_counts[needs_fill], ctx_counts[needs_fill]) / np.maximum(ctx_counts[needs_fill], 1.0) + best_p_ng[fill_idx] = np.clip(p, 0.0, 1.0) + + # Mix model probability with n-gram + has_match = best_p_ng >= 0 + if has_match.any(): + if use_entropy: + alpha = alpha_per_tok[has_match] + else: + alpha = base_alpha + seg_model_p[has_match] = (1.0 - alpha) * seg_model_p[has_match] + alpha * best_p_ng[has_match] + seg_nll = -np.log(np.clip(seg_model_p, 1e-12, 1.0)) + + # Score-first: update ALL order tables AFTER scoring + for oi in range(n_orders): + if order_data[oi] is None: + continue + v_idx, ctx_key, full_key = order_data[oi] + np.add.at(ctx_tables[oi], ctx_key, 1) + np.add.at(full_tables[oi], full_key, 1) + + loss_sum += float(seg_nll.sum()) + token_count += float(seg_len) + tgt = y_batch[i, s:wlen] + prev = x_batch[i, s:wlen] + tb = base_bytes_lut[tgt].to(torch.float64) + tb += (has_leading_space_lut[tgt] & ~is_boundary_token_lut[prev]).to(torch.float64) + byte_count += float(tb.sum().item()) + + if rank == 0 and (bi // batch_seqs) % 200 == 0 and bi > 0: + done = min(bi + batch_seqs, len(my_windows)) + pct = done / len(my_windows) * 100 + cur_bpb = (loss_sum / max(token_count, 1.0)) / math.log(2.0) * (token_count / max(byte_count, 1.0)) + elapsed = time.perf_counter() - eval_start + print(f" ngram_eval [{pct:5.1f}%] bpb={cur_bpb:.6f} t={elapsed:.0f}s", flush=True) + + _loss = torch.tensor(loss_sum, device=device, dtype=torch.float64) + _toks = torch.tensor(token_count, device=device, dtype=torch.float64) + _bytes = torch.tensor(byte_count, device=device, dtype=torch.float64) + if dist.is_available() and dist.is_initialized(): + dist.all_reduce(_loss, op=dist.ReduceOp.SUM) + dist.all_reduce(_toks, op=dist.ReduceOp.SUM) + dist.all_reduce(_bytes, op=dist.ReduceOp.SUM) + + val_loss = _loss.item() / max(_toks.item(), 1.0) + val_bpb = val_loss / math.log(2.0) * (_toks.item() / max(_bytes.item(), 1.0)) + # Coverage check: warn if eval was cut short + total_expected = sum(1 for ws in window_starts + if (min(ws + seq_len, total_tokens) - ws - (0 if ws == 0 else max(min(ws + seq_len, total_tokens) - ws - stride, 0))) > 0) + coverage = _toks.item() / max(total_expected * stride, 1.0) # approximate + elapsed = time.perf_counter() - eval_start + if rank == 0: + print(f" ngram_eval DONE: bpb={val_bpb:.6f} tokens={_toks.item():.0f} t={elapsed:.0f}s", flush=True) + if elapsed >= eval_budget_s - 10: + print(f" WARNING: eval used {elapsed:.0f}s of {eval_budget_s}s budget — results may be from partial coverage", flush=True) + base_model.train() + return val_loss, val_bpb + # ----------------------------- # TRAINING @@ -735,10 +1093,6 @@ def main() -> None: args = Hyperparameters() zeropower_via_newtonschulz5 = torch.compile(zeropower_via_newtonschulz5) - # ----------------------------- - # DISTRIBUTED + CUDA SETUP - # ----------------------------- - distributed = "RANK" in os.environ and "WORLD_SIZE" in os.environ rank = int(os.environ.get("RANK", "0")) world_size = int(os.environ.get("WORLD_SIZE", "1")) @@ -758,15 +1112,16 @@ def main() -> None: dist.barrier() master_process = rank == 0 - # Fast math knobs torch.backends.cuda.matmul.allow_tf32 = True torch.backends.cudnn.allow_tf32 = True - from torch.backends.cuda import enable_cudnn_sdp, enable_flash_sdp, enable_math_sdp, enable_mem_efficient_sdp - - enable_cudnn_sdp(False) - enable_flash_sdp(True) - enable_mem_efficient_sdp(False) - enable_math_sdp(False) + try: + from torch.backends.cuda import enable_cudnn_sdp, enable_flash_sdp, enable_math_sdp, enable_mem_efficient_sdp + enable_cudnn_sdp(False) + enable_flash_sdp(True) + enable_mem_efficient_sdp(False) + enable_math_sdp(False) + except ImportError: + pass logfile = None if master_process: @@ -793,10 +1148,6 @@ def log0(msg: str, console: bool = True) -> None: ) log0("=" * 100, console=False) - # ----------------------------- - # TOKENIZER + VALIDATION METRIC SETUP - # ----------------------------- - random.seed(args.seed) np.random.seed(args.seed) torch.manual_seed(args.seed) @@ -819,10 +1170,7 @@ def log0(msg: str, console: bool = True) -> None: log0(f"train_loader:dataset:{dataset_dir.name} train_shards:{actual_train_files}") log0(f"val_loader:shards pattern={args.val_files} tokens:{val_tokens.numel() - 1}") - # ----------------------------- # MODEL + OPTIMIZER SETUP - # ----------------------------- - base_model = GPT( vocab_size=args.vocab_size, num_layers=args.num_layers, @@ -835,37 +1183,47 @@ def log0(msg: str, console: bool = True) -> None: logit_softcap=args.logit_softcap, rope_base=args.rope_base, qk_gain_init=args.qk_gain_init, + bigram_vocab_size=args.bigram_vocab_size, + bigram_dim=args.bigram_dim, + rope_dims=args.rope_dims, + ln_scale=args.ln_scale, + xsa_last_n=args.xsa_last_n, ).to(device).bfloat16() for module in base_model.modules(): if isinstance(module, CastedLinear): module.float() restore_low_dim_params_to_fp32(base_model) + torch._dynamo.config.optimize_ddp = False compiled_model = torch.compile(base_model, dynamic=False, fullgraph=True) model: nn.Module = DDP(compiled_model, device_ids=[local_rank], broadcast_buffers=False) if distributed else compiled_model - # Optimizer split: - # - token embedding (Adam) uses EMBED_LR - # - untied lm_head (Adam) uses HEAD_LR - # - matrix params in transformer blocks use MATRIX_LR via Muon - # - vectors/scalars use SCALAR_LR via Adam block_named_params = list(base_model.blocks.named_parameters()) matrix_params = [ - p - for name, p in block_named_params + p for name, p in block_named_params if p.ndim == 2 and not any(pattern in name for pattern in CONTROL_TENSOR_NAME_PATTERNS) ] scalar_params = [ - p - for name, p in block_named_params + p for name, p in block_named_params if p.ndim < 2 or any(pattern in name for pattern in CONTROL_TENSOR_NAME_PATTERNS) ] if base_model.skip_weights.numel() > 0: scalar_params.append(base_model.skip_weights) + scalar_params.append(base_model.smear.gate) + if base_model.bigram is not None: + scalar_params.append(base_model.bigram.scale) + token_lr = args.tied_embed_lr if args.tie_embeddings else args.embed_lr - optimizer_tok = torch.optim.Adam( - [{"params": [base_model.tok_emb.weight], "lr": token_lr, "base_lr": token_lr}], + tok_params = [{"params": [base_model.tok_emb.weight], "lr": token_lr, "base_lr": token_lr}] + if base_model.bigram is not None: + tok_params.append({"params": [base_model.bigram.embed.weight], "lr": token_lr, "base_lr": token_lr}) + if base_model.bigram.proj is not None: + matrix_params.append(base_model.bigram.proj.weight) + + optimizer_tok = torch.optim.AdamW( + tok_params, betas=(args.beta1, args.beta2), eps=args.adam_eps, + weight_decay=args.weight_decay, fused=True, ) optimizer_muon = Muon( @@ -873,13 +1231,15 @@ def log0(msg: str, console: bool = True) -> None: lr=args.matrix_lr, momentum=args.muon_momentum, backend_steps=args.muon_backend_steps, + weight_decay=0.04, ) for group in optimizer_muon.param_groups: group["base_lr"] = args.matrix_lr - optimizer_scalar = torch.optim.Adam( + optimizer_scalar = torch.optim.AdamW( [{"params": scalar_params, "lr": args.scalar_lr, "base_lr": args.scalar_lr}], betas=(args.beta1, args.beta2), eps=args.adam_eps, + weight_decay=args.weight_decay, fused=True, ) optimizers: list[torch.optim.Optimizer] = [optimizer_tok, optimizer_muon, optimizer_scalar] @@ -895,11 +1255,9 @@ def log0(msg: str, console: bool = True) -> None: n_params = sum(p.numel() for p in base_model.parameters()) log0(f"model_params:{n_params}") log0(f"world_size:{world_size} grad_accum_steps:{grad_accum_steps}") - log0("sdp_backends:cudnn=False flash=True mem_efficient=False math=False") log0(f"attention_mode:gqa num_heads:{args.num_heads} num_kv_heads:{args.num_kv_heads}") log0( f"tie_embeddings:{args.tie_embeddings} embed_lr:{token_lr} " - f"head_lr:{args.head_lr if base_model.lm_head is not None else 0.0} " f"matrix_lr:{args.matrix_lr} scalar_lr:{args.scalar_lr}" ) log0( @@ -909,10 +1267,7 @@ def log0(msg: str, console: bool = True) -> None: ) log0(f"seed:{args.seed}") - # ----------------------------- # DATA LOADER & MODEL WARMUP - # ----------------------------- - train_loader = DistributedTokenLoader(args.train_files, rank, world_size, device) def zero_grad_all() -> None: @@ -932,8 +1287,6 @@ def lr_mul(step: int, elapsed_ms: float) -> float: remaining_ms = max(max_wallclock_ms - elapsed_ms, 0.0) return remaining_ms / max(warmdown_ms, 1e-9) if remaining_ms <= warmdown_ms else 1.0 - # Warmup primes the compiled forward/backward/optimizer paths, then we restore the - # initial weights/optimizer state so measured training starts from the true init. if args.warmup_steps > 0: initial_model_state = {name: tensor.detach().cpu().clone() for name, tensor in base_model.state_dict().items()} initial_optimizer_states = [copy.deepcopy(opt.state_dict()) for opt in optimizers] @@ -960,12 +1313,16 @@ def lr_mul(step: int, elapsed_ms: float) -> float: model.require_backward_grad_sync = True train_loader = DistributedTokenLoader(args.train_files, rank, world_size, device) - # ----------------------------- - # MAIN TRAINING LOOP - # ----------------------------- + # EMA shadow model (kept on GPU to avoid PCIe bottleneck) + ema_state: dict[str, Tensor] | None = None + if args.ema_enabled: + ema_state = {name: t.detach().clone() for name, t in base_model.state_dict().items()} + # MAIN TRAINING LOOP training_time_ms = 0.0 stop_after_step: int | None = None + swa_state: dict[str, Tensor] | None = None + swa_count = 0 torch.cuda.synchronize() t0 = time.perf_counter() @@ -978,16 +1335,8 @@ def lr_mul(step: int, elapsed_ms: float) -> float: torch.cuda.synchronize() training_time_ms += 1000.0 * (time.perf_counter() - t0) val_loss, val_bpb = eval_val( - args, - model, - rank, - world_size, - device, - grad_accum_steps, - val_tokens, - base_bytes_lut, - has_leading_space_lut, - is_boundary_token_lut, + args, model, rank, world_size, device, grad_accum_steps, + val_tokens, base_bytes_lut, has_leading_space_lut, is_boundary_token_lut, ) log0( f"step:{step}/{args.iterations} val_loss:{val_loss:.4f} val_bpb:{val_bpb:.4f} " @@ -1034,7 +1383,27 @@ def lr_mul(step: int, elapsed_ms: float) -> float: zero_grad_all() step += 1 + + # EMA update every 10 steps (GPU-resident, amortize overhead) + if ema_state is not None and step % 10 == 0: + decay = args.ema_decay ** 10 # compensate for batched updates + with torch.no_grad(): + for name, param in base_model.state_dict().items(): + ema_state[name].lerp_(param.detach(), 1.0 - decay) + approx_training_time_ms = training_time_ms + 1000.0 * (time.perf_counter() - t0) + + # SWA: collect checkpoints during warmdown + if args.swa_enabled and scale < args.swa_start_frac and step % args.swa_every == 0: + if swa_state is None: + swa_state = {name: t.detach().cpu().clone() for name, t in base_model.state_dict().items()} + swa_count = 1 + log0(f"swa:start step:{step}") + else: + for name, t in base_model.state_dict().items(): + swa_state[name] += t.detach().cpu() + swa_count += 1 + should_log_train = ( args.train_log_every > 0 and (step <= 10 or step % args.train_log_every == 0 or stop_after_step is not None) @@ -1045,7 +1414,6 @@ def lr_mul(step: int, elapsed_ms: float) -> float: f"train_time:{approx_training_time_ms:.0f}ms step_avg:{approx_training_time_ms / step:.2f}ms" ) - # Needed to sync whether we've reached the wallclock cap. reached_cap = max_wallclock_ms is not None and approx_training_time_ms >= max_wallclock_ms if distributed and max_wallclock_ms is not None: reached_cap_tensor = torch.tensor(int(reached_cap), device=device) @@ -1059,12 +1427,27 @@ def lr_mul(step: int, elapsed_ms: float) -> float: f"reserved: {torch.cuda.max_memory_reserved() // 1024 // 1024} MiB" ) - # ----------------------------- - # SERIALIZATION + ROUNDTRIP VALIDATION - # ----------------------------- - # Save the raw state (useful for debugging/loading in PyTorch directly), then always produce - # the compressed int8+zlib artifact and validate the round-tripped weights. + # Apply SWA if collected + if args.swa_enabled and swa_state is not None and swa_count > 1: + log0(f"swa:applying averaged {swa_count} checkpoints") + current_state = base_model.state_dict() + avg_state = { + name: (tensor / swa_count).to(dtype=current_state[name].dtype) + for name, tensor in swa_state.items() + } + base_model.load_state_dict(avg_state, strict=True) + + # Apply EMA if enabled (overrides SWA) + if args.ema_enabled and ema_state is not None: + log0("ema:applying shadow model") + current_state = base_model.state_dict() + ema_applied = { + name: tensor.to(dtype=current_state[name].dtype, device=current_state[name].device) + for name, tensor in ema_state.items() + } + base_model.load_state_dict(ema_applied, strict=True) + # SERIALIZATION + ROUNDTRIP VALIDATION if master_process: torch.save(base_model.state_dict(), "final_model.pt") model_bytes = os.path.getsize("final_model.pt") @@ -1073,44 +1456,74 @@ def lr_mul(step: int, elapsed_ms: float) -> float: log0(f"Code size: {code_bytes} bytes") log0(f"Total submission size: {model_bytes + code_bytes} bytes") - quant_obj, quant_stats = quantize_state_dict_int8(base_model.state_dict()) + # Magnitude pruning: zero out smallest weights to improve compression + with torch.no_grad(): + for name, param in base_model.named_parameters(): + if param.ndim == 2 and param.numel() > 65536: + threshold = torch.quantile(param.abs().float().flatten(), 0.03) + mask = param.abs() < threshold + param.masked_fill_(mask, 0.0) + + # INT6 mixed quantization + zstd/zlib export + sd_cpu = {k: v.detach().cpu() for k, v in base_model.state_dict().items()} + quant_result, quant_meta = mixed_quantize_int6(sd_cpu, {"mlp", "attn", "bigram"}) quant_buf = io.BytesIO() - torch.save(quant_obj, quant_buf) + torch.save({"w": quant_result, "m": quant_meta}, quant_buf) quant_raw = quant_buf.getvalue() - quant_blob = zlib.compress(quant_raw, level=9) - quant_raw_bytes = len(quant_raw) + if _COMPRESSOR == "zstd": + quant_blob = zstandard.ZstdCompressor(level=22).compress(quant_raw) + else: + quant_blob = zlib.compress(quant_raw, 9) if master_process: with open("final_model.int8.ptz", "wb") as f: f.write(quant_blob) quant_file_bytes = os.path.getsize("final_model.int8.ptz") code_bytes = len(code.encode("utf-8")) - ratio = quant_stats["baseline_tensor_bytes"] / max(quant_stats["int8_payload_bytes"], 1) - log0( - f"Serialized model int8+zlib: {quant_file_bytes} bytes " - f"(payload:{quant_stats['int8_payload_bytes']} raw_torch:{quant_raw_bytes} payload_ratio:{ratio:.2f}x)" - ) - log0(f"Total submission size int8+zlib: {quant_file_bytes + code_bytes} bytes") + log0(f"Serialized model int6+{_COMPRESSOR}: {quant_file_bytes} bytes") + total_bytes = quant_file_bytes + code_bytes + log0(f"Total submission size: {total_bytes} bytes ({total_bytes/1e6:.2f} MB)") + if total_bytes > 16_000_000: + log0(f"FAILSAFE: artifact {total_bytes} bytes EXCEEDS 16MB limit! Aborting eval.") + sys.exit(1) + log0(f"SIZE CHECK PASSED: {total_bytes/1e6:.2f} MB < 16.00 MB") if distributed: dist.barrier() with open("final_model.int8.ptz", "rb") as f: quant_blob_disk = f.read() - quant_state = torch.load(io.BytesIO(zlib.decompress(quant_blob_disk)), map_location="cpu") - base_model.load_state_dict(dequantize_state_dict_int8(quant_state), strict=True) + if _COMPRESSOR == "zstd": + decompressed = zstandard.ZstdDecompressor().decompress(quant_blob_disk) + else: + decompressed = zlib.decompress(quant_blob_disk) + quant_state = torch.load(io.BytesIO(decompressed), map_location="cpu") + deq_state = dequantize_mixed_int6(quant_state["w"], quant_state["m"], sd_cpu) + base_model.load_state_dict(deq_state, strict=True) + + # Sliding window eval on int6-roundtripped weights torch.cuda.synchronize() t_qeval = time.perf_counter() - q_val_loss, q_val_bpb = eval_val( - args, - model, - rank, - world_size, - device, - grad_accum_steps, - val_tokens, - base_bytes_lut, - has_leading_space_lut, - is_boundary_token_lut, - ) + if args.ngram_eval_max_order >= 2 and args.eval_stride > 0: + log0(f"final_eval_mode:sliding_ngram orders={args.ngram_eval_min_order}-{args.ngram_eval_max_order} " + f"alpha={args.ngram_eval_alpha} entropy={args.ngram_eval_entropy} stride:{args.eval_stride}") + q_val_loss, q_val_bpb = eval_val_sliding_ngram( + args, base_model, rank, world_size, device, + val_tokens, base_bytes_lut, has_leading_space_lut, is_boundary_token_lut, + stride=args.eval_stride, + batch_seqs=args.eval_batch_seqs, + ) + elif args.eval_stride > 0 and args.eval_stride < args.train_seq_len: + log0(f"final_eval_mode:sliding_window stride:{args.eval_stride} batch_seqs:{args.eval_batch_seqs}") + q_val_loss, q_val_bpb = eval_val_sliding( + args, base_model, rank, world_size, device, + val_tokens, base_bytes_lut, has_leading_space_lut, is_boundary_token_lut, + stride=args.eval_stride, batch_seqs=args.eval_batch_seqs, + ) + else: + log0("final_eval_mode:standard") + q_val_loss, q_val_bpb = eval_val( + args, model, rank, world_size, device, grad_accum_steps, + val_tokens, base_bytes_lut, has_leading_space_lut, is_boundary_token_lut, + ) torch.cuda.synchronize() log0( f"final_int8_zlib_roundtrip val_loss:{q_val_loss:.4f} val_bpb:{q_val_bpb:.4f} " @@ -1124,3 +1537,5 @@ def lr_mul(step: int, elapsed_ms: float) -> float: if __name__ == "__main__": main() +# fixes applied +# tuned