CodeGen Architecture Adapter

tests/unit/model_bridge/supported_architectures/test_codegen_adapter.py

@@ -0,0 +1,328 @@
+"""Unit tests for CodeGenArchitectureAdapter.
+
+Tests cover:
+- Config attribute validation (all required attributes are set correctly)
+- Component mapping structure (correct bridge types, no ln2)
+- Weight conversion keys and structure
+- split_qkv_matrix correctness (numerical test with known weights)
+- Factory registration (CodeGenForCausalLM maps to the right adapter)
+"""
+
+from types import SimpleNamespace
+from typing import Any
+
+import pytest
+import torch
+import torch.nn as nn
+
+from transformer_lens.config import TransformerBridgeConfig
+from transformer_lens.model_bridge.generalized_components import (
+    BlockBridge,
+    CodeGenAttentionBridge,
+    EmbeddingBridge,
+    MLPBridge,
+    NormalizationBridge,
+    UnembeddingBridge,
+)
+from transformer_lens.model_bridge.supported_architectures.codegen import (
+    CodeGenArchitectureAdapter,
+)
+
+# ---------------------------------------------------------------------------
+# Fixtures
+# ---------------------------------------------------------------------------
+
+
+def _make_cfg(
+    n_heads: int = 4,
+    d_model: int = 64,
+    n_layers: int = 2,
+    d_mlp: int = 256,
+    d_vocab: int = 1000,
+    n_ctx: int = 512,
+) -> TransformerBridgeConfig:
+    """Return a minimal TransformerBridgeConfig for CodeGen adapter tests."""
+    return TransformerBridgeConfig(
+        d_model=d_model,
+        d_head=d_model // n_heads,
+        n_layers=n_layers,
+        n_ctx=n_ctx,
+        n_heads=n_heads,
+        d_vocab=d_vocab,
+        d_mlp=d_mlp,
+        default_prepend_bos=True,
+        architecture="CodeGenForCausalLM",
+    )
+
+
+@pytest.fixture
+def cfg() -> TransformerBridgeConfig:
+    return _make_cfg()
+
+
+@pytest.fixture
+def adapter(cfg: TransformerBridgeConfig) -> CodeGenArchitectureAdapter:
+    return CodeGenArchitectureAdapter(cfg)
+
+
+# ---------------------------------------------------------------------------
+# Config attribute tests
+# ---------------------------------------------------------------------------
+
+
+class TestCodeGenAdapterConfig:
+    """Tests that the adapter sets required config attributes correctly."""
+
+    def test_normalization_type_is_ln(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert adapter.cfg.normalization_type == "LN"
+
+    def test_positional_embedding_type_is_rotary(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert adapter.cfg.positional_embedding_type == "rotary"
+
+    def test_final_rms_is_false(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert adapter.cfg.final_rms is False
+
+    def test_gated_mlp_is_false(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert adapter.cfg.gated_mlp is False
+
+    def test_attn_only_is_false(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert adapter.cfg.attn_only is False
+
+    def test_parallel_attn_mlp_is_true(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert adapter.cfg.parallel_attn_mlp is True
+
+
+# ---------------------------------------------------------------------------
+# Component mapping structure tests
+# ---------------------------------------------------------------------------
+
+
+class TestCodeGenAdapterComponentMapping:
+    """Tests that component_mapping has the correct bridge types and structure."""
+
+    def test_embed_is_embedding_bridge(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert isinstance(adapter.component_mapping["embed"], EmbeddingBridge)
+
+    def test_embed_name(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert adapter.component_mapping["embed"].name == "transformer.wte"
+
+    def test_blocks_is_block_bridge(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert isinstance(adapter.component_mapping["blocks"], BlockBridge)
+
+    def test_blocks_name(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert adapter.component_mapping["blocks"].name == "transformer.h"
+
+    def test_ln_final_is_normalization_bridge(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert isinstance(adapter.component_mapping["ln_final"], NormalizationBridge)
+
+    def test_ln_final_name(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert adapter.component_mapping["ln_final"].name == "transformer.ln_f"
+
+    def test_unembed_is_unembedding_bridge(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert isinstance(adapter.component_mapping["unembed"], UnembeddingBridge)
+
+    def test_unembed_name(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert adapter.component_mapping["unembed"].name == "lm_head"
+
+    def test_blocks_ln1_is_normalization_bridge(self, adapter: CodeGenArchitectureAdapter) -> None:
+        blocks = adapter.component_mapping["blocks"]
+        assert isinstance(blocks.submodules["ln1"], NormalizationBridge)
+
+    def test_blocks_ln1_name(self, adapter: CodeGenArchitectureAdapter) -> None:
+        blocks = adapter.component_mapping["blocks"]
+        assert blocks.submodules["ln1"].name == "ln_1"
+
+    def test_no_ln2_in_blocks(self, adapter: CodeGenArchitectureAdapter) -> None:
+        """CodeGen uses parallel attn+MLP sharing ln_1 — there must be no ln2."""
+        blocks = adapter.component_mapping["blocks"]
+        assert "ln2" not in blocks.submodules, "CodeGen parallel block must not have ln2"
+
+    def test_attn_is_codegen_attention_bridge(self, adapter: CodeGenArchitectureAdapter) -> None:
+        blocks = adapter.component_mapping["blocks"]
+        assert isinstance(blocks.submodules["attn"], CodeGenAttentionBridge)
+
+    def test_attn_name(self, adapter: CodeGenArchitectureAdapter) -> None:
+        blocks = adapter.component_mapping["blocks"]
+        assert blocks.submodules["attn"].name == "attn"
+
+    def test_mlp_is_mlp_bridge(self, adapter: CodeGenArchitectureAdapter) -> None:
+        blocks = adapter.component_mapping["blocks"]
+        assert isinstance(blocks.submodules["mlp"], MLPBridge)
+
+    def test_mlp_name(self, adapter: CodeGenArchitectureAdapter) -> None:
+        blocks = adapter.component_mapping["blocks"]
+        assert blocks.submodules["mlp"].name == "mlp"
+
+    def test_mlp_in_name(self, adapter: CodeGenArchitectureAdapter) -> None:
+        blocks = adapter.component_mapping["blocks"]
+        assert blocks.submodules["mlp"].submodules["in"].name == "fc_in"
+
+    def test_mlp_out_name(self, adapter: CodeGenArchitectureAdapter) -> None:
+        blocks = adapter.component_mapping["blocks"]
+        assert blocks.submodules["mlp"].submodules["out"].name == "fc_out"
+
+
+# ---------------------------------------------------------------------------
+# Weight processing conversion tests
+# ---------------------------------------------------------------------------
+
+
+class TestCodeGenAdapterWeightConversions:
+    """Tests that weight_processing_conversions has the expected keys."""
+
+    def test_q_weight_key_present(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert "blocks.{i}.attn.q.weight" in adapter.weight_processing_conversions
+
+    def test_k_weight_key_present(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert "blocks.{i}.attn.k.weight" in adapter.weight_processing_conversions
+
+    def test_v_weight_key_present(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert "blocks.{i}.attn.v.weight" in adapter.weight_processing_conversions
+
+    def test_o_weight_key_present(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert "blocks.{i}.attn.o.weight" in adapter.weight_processing_conversions
+
+    def test_exactly_four_conversion_keys(self, adapter: CodeGenArchitectureAdapter) -> None:
+        assert len(adapter.weight_processing_conversions) == 4
+
+
+# ---------------------------------------------------------------------------
+# split_qkv_matrix numerical correctness tests
+# ---------------------------------------------------------------------------
+
+
+class TestCodeGenSplitQKVMatrix:
+    """Numerical tests verifying the mp_num=4 QKV split logic."""
+
+    def _make_adapter_with_dmodel(self, d_model: int, n_heads: int) -> CodeGenArchitectureAdapter:
+        cfg = _make_cfg(d_model=d_model, n_heads=n_heads)
+        return CodeGenArchitectureAdapter(cfg)
+
+    def _make_attn_component(self, d_model: int) -> Any:
+        """Create a minimal attn component with a qkv_proj linear."""
+        attn = SimpleNamespace()
+        attn.qkv_proj = nn.Linear(d_model, d_model * 3, bias=False)
+        return attn
+
+    def test_returns_three_linear_modules(self) -> None:
+        """split_qkv_matrix must return exactly three nn.Linear modules."""
+        adapter = self._make_adapter_with_dmodel(64, 4)
+        attn = self._make_attn_component(64)
+        q, k, v = adapter.split_qkv_matrix(attn)
+        assert isinstance(q, nn.Linear)
+        assert isinstance(k, nn.Linear)
+        assert isinstance(v, nn.Linear)
+
+    def test_output_shapes_are_correct(self) -> None:
+        """Each of Q, K, V must have weight shape [n_embd, n_embd]."""
+        d_model = 64
+        adapter = self._make_adapter_with_dmodel(d_model, 4)
+        attn = self._make_attn_component(d_model)
+        q, k, v = adapter.split_qkv_matrix(attn)
+        assert q.weight.shape == (d_model, d_model)
+        assert k.weight.shape == (d_model, d_model)
+        assert v.weight.shape == (d_model, d_model)
+
+    def test_no_bias_on_outputs(self) -> None:
+        """The split linears must have no bias, matching qkv_proj."""
+        adapter = self._make_adapter_with_dmodel(64, 4)
+        attn = self._make_attn_component(64)
+        q, k, v = adapter.split_qkv_matrix(attn)
+        assert q.bias is None
+        assert k.bias is None
+        assert v.bias is None
+
+    def test_q_k_v_are_distinct(self) -> None:
+        """With a non-trivial weight, Q, K, V must differ from each other."""
+        adapter = self._make_adapter_with_dmodel(64, 4)
+        attn = self._make_attn_component(64)
+        # Fill qkv_proj with distinct values per row
+        nn.init.normal_(attn.qkv_proj.weight)
+        q, k, v = adapter.split_qkv_matrix(attn)
+        # All three must differ
+        assert not torch.allclose(q.weight, k.weight), "Q and K weights must differ"
+        assert not torch.allclose(q.weight, v.weight), "Q and V weights must differ"
+        assert not torch.allclose(k.weight, v.weight), "K and V weights must differ"
+
+    def test_known_partition_ordering(self) -> None:
+        """Verify the mp_num=4 partition layout: within each partition [Q_part, V_part, K_part].
+
+        We construct a weight where partition index and slot index are embedded
+        in the values, then verify that Q, K, V extract the correct slices.
+        """
+        mp_num = 4
+        d_model = 64
+        n_heads = 4
+        local_dim = d_model // mp_num  # 16
+
+        adapter = self._make_adapter_with_dmodel(d_model, n_heads)
+        attn = self._make_attn_component(d_model)
+
+        # Build a structured weight: rows are indexed 0..3*d_model-1.
+        # Reshape as [mp_num=4, 3, local_dim=16, d_model=64], set each slice
+        # to a unique constant so we can track which slot goes where.
+        w = torch.zeros(mp_num, 3, local_dim, d_model)
+        # slot 0 = Q_part → fill with 1.0
+        w[:, 0, :, :] = 1.0
+        # slot 1 = V_part → fill with 2.0
+        w[:, 1, :, :] = 2.0
+        # slot 2 = K_part → fill with 3.0
+        w[:, 2, :, :] = 3.0
+
+        # Flatten back to [3*d_model, d_model] as qkv_proj expects
+        attn.qkv_proj.weight = nn.Parameter(w.reshape(3 * d_model, d_model))
+
+        q, k, v = adapter.split_qkv_matrix(attn)
+
+        assert torch.all(q.weight == 1.0), "Q should come from slot 0 (Q_part)"
+        assert torch.all(k.weight == 3.0), "K should come from slot 2 (K_part)"
+        assert torch.all(v.weight == 2.0), "V should come from slot 1 (V_part)"
+
+    def test_forward_output_shape_with_split(self) -> None:
+        """After split, Q/K/V linears should produce correct output shapes."""
+        d_model = 64
+        adapter = self._make_adapter_with_dmodel(d_model, 4)
+        attn = self._make_attn_component(d_model)
+        q_lin, k_lin, v_lin = adapter.split_qkv_matrix(attn)
+
+        batch, seq = 2, 10
+        x = torch.randn(batch, seq, d_model)
+        assert q_lin(x).shape == (batch, seq, d_model)
+        assert k_lin(x).shape == (batch, seq, d_model)
+        assert v_lin(x).shape == (batch, seq, d_model)
+
+
+# ---------------------------------------------------------------------------
+# Factory registration test
+# ---------------------------------------------------------------------------
+
+
+class TestCodeGenFactoryRegistration:
+    """Tests that the factory maps CodeGenForCausalLM to the correct adapter.
+
+    Note: Phase D (registration) is required for these tests to pass.  They
+    are included here so that registration is verified as part of the Phase D
+    commit rather than needing a separate test file.
+    """
+
+    def test_factory_returns_codegen_adapter(self) -> None:
+        """ArchitectureAdapterFactory must return a CodeGenArchitectureAdapter."""
+        from transformer_lens.factories.architecture_adapter_factory import (
+            ArchitectureAdapterFactory,
+        )
+
+        cfg = _make_cfg()
+        adapter = ArchitectureAdapterFactory.select_architecture_adapter(cfg)
+        assert isinstance(
+            adapter, CodeGenArchitectureAdapter
+        ), f"Expected CodeGenArchitectureAdapter, got {type(adapter).__name__}"
+
+    def test_factory_key_is_codegen_for_causal_lm(self) -> None:
+        """SUPPORTED_ARCHITECTURES must have a 'CodeGenForCausalLM' key."""
+        from transformer_lens.factories.architecture_adapter_factory import (
+            SUPPORTED_ARCHITECTURES,
+        )
+
+        assert (
+            "CodeGenForCausalLM" in SUPPORTED_ARCHITECTURES
+        ), "CodeGenForCausalLM must be registered in SUPPORTED_ARCHITECTURES"

transformer_lens/factories/architecture_adapter_factory.py

@@ -9,6 +9,7 @@
     ApertusArchitectureAdapter,
     BertArchitectureAdapter,
     BloomArchitectureAdapter,
+    CodeGenArchitectureAdapter,
     FalconArchitectureAdapter,
     Gemma1ArchitectureAdapter,
     Gemma2ArchitectureAdapter,
@@ -53,6 +54,7 @@
     "ApertusForCausalLM": ApertusArchitectureAdapter,
     "BertForMaskedLM": BertArchitectureAdapter,
     "BloomForCausalLM": BloomArchitectureAdapter,
+    "CodeGenForCausalLM": CodeGenArchitectureAdapter,
     "FalconForCausalLM": FalconArchitectureAdapter,
     "GemmaForCausalLM": Gemma1ArchitectureAdapter,  # Default to Gemma1 as it's the original version
     "Gemma1ForCausalLM": Gemma1ArchitectureAdapter,

transformer_lens/model_bridge/generalized_components/__init__.py

@@ -9,6 +9,9 @@
 from transformer_lens.model_bridge.generalized_components.bloom_attention import (
     BloomAttentionBridge,
 )
+from transformer_lens.model_bridge.generalized_components.codegen_attention import (
+    CodeGenAttentionBridge,
+)
 from transformer_lens.model_bridge.generalized_components.bloom_block import (
     BloomBlockBridge,
 )
@@ -78,6 +81,7 @@
     "BlockBridge",
     "BloomBlockBridge",
     "BloomAttentionBridge",
+    "CodeGenAttentionBridge",
     "BloomMLPBridge",
     "CLIPVisionEncoderBridge",
     "CLIPVisionEncoderLayerBridge",