| transformer_lens/ActivationCache.py |
461 |
18 |
2 |
240 |
20 |
94% |
| transformer_lens/BertNextSentencePrediction.py |
61 |
1 |
38 |
14 |
2 |
96% |
| transformer_lens/FactoredMatrix.py |
135 |
3 |
58 |
28 |
2 |
97% |
| transformer_lens/HookedAudioEncoder.py |
210 |
39 |
12 |
66 |
20 |
73% |
| transformer_lens/HookedEncoder.py |
192 |
58 |
40 |
44 |
5 |
62% |
| transformer_lens/HookedEncoderDecoder.py |
277 |
173 |
32 |
100 |
0 |
28% |
| transformer_lens/HookedRootModule.py |
178 |
40 |
0 |
88 |
13 |
73% |
| transformer_lens/HookedTransformer.py |
832 |
229 |
80 |
414 |
55 |
68% |
| transformer_lens/SVDInterpreter.py |
55 |
1 |
0 |
14 |
1 |
97% |
| transformer_lens/__init__.py |
24 |
2 |
0 |
2 |
1 |
88% |
| transformer_lens/benchmarks/__init__.py |
11 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/benchmarks/activation_cache.py |
69 |
59 |
0 |
36 |
0 |
10% |
| transformer_lens/benchmarks/audio.py |
127 |
94 |
0 |
48 |
2 |
22% |
| transformer_lens/benchmarks/backward_gradients.py |
196 |
184 |
0 |
102 |
0 |
4% |
| transformer_lens/benchmarks/component_benchmark.py |
34 |
29 |
0 |
18 |
0 |
10% |
| transformer_lens/benchmarks/component_outputs.py |
461 |
162 |
0 |
218 |
45 |
61% |
| transformer_lens/benchmarks/encoder_common.py |
87 |
36 |
0 |
34 |
7 |
55% |
| transformer_lens/benchmarks/forward_pass.py |
111 |
96 |
0 |
56 |
0 |
9% |
| transformer_lens/benchmarks/generation.py |
65 |
30 |
0 |
30 |
5 |
48% |
| transformer_lens/benchmarks/granular_weight_processing.py |
161 |
161 |
0 |
84 |
0 |
0% |
| transformer_lens/benchmarks/hook_registration.py |
237 |
175 |
0 |
112 |
4 |
24% |
| transformer_lens/benchmarks/hook_structure.py |
149 |
137 |
0 |
80 |
0 |
5% |
| transformer_lens/benchmarks/main_benchmark.py |
1027 |
983 |
0 |
574 |
1 |
3% |
| transformer_lens/benchmarks/multimodal.py |
106 |
63 |
0 |
48 |
4 |
37% |
| transformer_lens/benchmarks/text_quality.py |
160 |
24 |
0 |
56 |
19 |
79% |
| transformer_lens/benchmarks/utils.py |
183 |
68 |
0 |
68 |
11 |
60% |
| transformer_lens/benchmarks/vision.py |
94 |
68 |
0 |
38 |
1 |
23% |
| transformer_lens/benchmarks/weight_processing.py |
326 |
301 |
0 |
158 |
0 |
5% |
| transformer_lens/cache/__init__.py |
1 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/cache/key_value_cache.py |
37 |
4 |
2 |
10 |
1 |
85% |
| transformer_lens/cache/key_value_cache_entry.py |
21 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/components/__init__.py |
22 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/components/abstract_attention.py |
372 |
67 |
0 |
130 |
31 |
78% |
| transformer_lens/components/attention.py |
20 |
4 |
0 |
4 |
2 |
75% |
| transformer_lens/components/bert_block.py |
45 |
4 |
0 |
2 |
1 |
89% |
| transformer_lens/components/bert_embed.py |
30 |
9 |
0 |
2 |
0 |
66% |
| transformer_lens/components/bert_mlm_head.py |
19 |
3 |
0 |
0 |
0 |
84% |
| transformer_lens/components/bert_nsp_head.py |
16 |
1 |
0 |
0 |
0 |
94% |
| transformer_lens/components/bert_pooler.py |
19 |
3 |
0 |
0 |
0 |
84% |
| transformer_lens/components/embed.py |
17 |
1 |
0 |
4 |
2 |
86% |
| transformer_lens/components/grouped_query_attention.py |
66 |
7 |
0 |
6 |
1 |
89% |
| transformer_lens/components/layer_norm.py |
25 |
1 |
0 |
4 |
1 |
93% |
| transformer_lens/components/layer_norm_pre.py |
19 |
1 |
0 |
2 |
1 |
90% |
| transformer_lens/components/mlps/can_be_used_as_mlp.py |
33 |
0 |
0 |
6 |
0 |
100% |
| transformer_lens/components/mlps/gated_mlp.py |
33 |
2 |
0 |
6 |
2 |
90% |
| transformer_lens/components/mlps/gated_mlp_4bit.py |
35 |
18 |
0 |
4 |
1 |
46% |
| transformer_lens/components/mlps/gpt_oss_moe.py |
62 |
0 |
0 |
4 |
0 |
100% |
| transformer_lens/components/mlps/mlp.py |
25 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/components/mlps/moe.py |
60 |
2 |
0 |
6 |
2 |
94% |
| transformer_lens/components/pos_embed.py |
23 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/components/rms_norm.py |
26 |
2 |
0 |
6 |
2 |
88% |
| transformer_lens/components/rms_norm_pre.py |
18 |
7 |
0 |
2 |
0 |
55% |
| transformer_lens/components/t5_attention.py |
50 |
34 |
0 |
10 |
0 |
27% |
| transformer_lens/components/t5_block.py |
64 |
49 |
0 |
12 |
0 |
20% |
| transformer_lens/components/token_typed_embed.py |
12 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/components/transformer_block.py |
110 |
13 |
0 |
56 |
8 |
84% |
| transformer_lens/components/unembed.py |
19 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/config/__init__.py |
4 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/config/hooked_transformer_config.py |
137 |
7 |
0 |
32 |
5 |
93% |
| transformer_lens/config/transformer_bridge_config.py |
86 |
1 |
0 |
4 |
2 |
97% |
| transformer_lens/config/transformer_lens_config.py |
62 |
3 |
0 |
12 |
2 |
93% |
| transformer_lens/conversion_utils/__init__.py |
2 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/conversion_utils/conversion_steps/__init__.py |
13 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/conversion_utils/conversion_steps/arithmetic_tensor_conversion.py |
39 |
11 |
0 |
16 |
1 |
64% |
| transformer_lens/conversion_utils/conversion_steps/attention_auto_conversion.py |
51 |
25 |
0 |
22 |
2 |
38% |
| transformer_lens/conversion_utils/conversion_steps/base_tensor_conversion.py |
14 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/conversion_utils/conversion_steps/callable_tensor_conversion.py |
10 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/conversion_utils/conversion_steps/chain_tensor_conversion.py |
15 |
4 |
0 |
4 |
0 |
58% |
| transformer_lens/conversion_utils/conversion_steps/rearrange_tensor_conversion.py |
18 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/conversion_utils/conversion_steps/repeat_tensor_conversion.py |
14 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/conversion_utils/conversion_steps/split_tensor_conversion.py |
12 |
1 |
0 |
0 |
0 |
92% |
| transformer_lens/conversion_utils/conversion_steps/tensor_conversion_set.py |
53 |
17 |
0 |
18 |
1 |
61% |
| transformer_lens/conversion_utils/conversion_steps/ternary_tensor_conversion.py |
37 |
0 |
0 |
14 |
1 |
98% |
| transformer_lens/conversion_utils/conversion_steps/transpose_tensor_conversion.py |
17 |
4 |
0 |
8 |
4 |
68% |
| transformer_lens/conversion_utils/conversion_steps/zeros_like_conversion.py |
7 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/conversion_utils/helpers/__init__.py |
3 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/conversion_utils/helpers/find_property.py |
7 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/conversion_utils/helpers/merge_quantiziation_fields.py |
17 |
1 |
0 |
10 |
1 |
93% |
| transformer_lens/conversion_utils/hook_conversion_utils.py |
23 |
7 |
0 |
10 |
1 |
70% |
| transformer_lens/conversion_utils/param_processing_conversion.py |
22 |
1 |
0 |
4 |
2 |
88% |
| transformer_lens/evals.py |
223 |
59 |
0 |
56 |
6 |
72% |
| transformer_lens/factories/__init__.py |
0 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/factories/activation_function_factory.py |
14 |
2 |
0 |
6 |
2 |
80% |
| transformer_lens/factories/architecture_adapter_factory.py |
38 |
2 |
0 |
10 |
0 |
96% |
| transformer_lens/factories/mlp_factory.py |
17 |
0 |
0 |
6 |
0 |
100% |
| transformer_lens/head_detector.py |
84 |
2 |
0 |
28 |
2 |
96% |
| transformer_lens/hook_points.py |
198 |
22 |
5 |
84 |
9 |
86% |
| transformer_lens/lit/__init__.py |
61 |
36 |
0 |
24 |
1 |
31% |
| transformer_lens/lit/constants.py |
80 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/lit/dataset.py |
187 |
82 |
3 |
26 |
2 |
50% |
| transformer_lens/lit/model.py |
227 |
162 |
3 |
66 |
3 |
23% |
| transformer_lens/lit/utils.py |
122 |
50 |
0 |
54 |
6 |
55% |
| transformer_lens/loading_from_pretrained.py |
497 |
182 |
0 |
328 |
78 |
59% |
| transformer_lens/model_bridge/__init__.py |
7 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/architecture_adapter.py |
504 |
111 |
0 |
274 |
52 |
75% |
| transformer_lens/model_bridge/bridge.py |
1931 |
377 |
3 |
1058 |
157 |
77% |
| transformer_lens/model_bridge/buffer_restore.py |
33 |
3 |
0 |
16 |
1 |
92% |
| transformer_lens/model_bridge/compat.py |
22 |
8 |
0 |
10 |
0 |
62% |
| transformer_lens/model_bridge/component_setup.py |
156 |
14 |
2 |
84 |
7 |
89% |
| transformer_lens/model_bridge/composition_scores.py |
49 |
7 |
0 |
6 |
3 |
82% |
| transformer_lens/model_bridge/exceptions.py |
6 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/generalized_components/__init__.py |
49 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/generalized_components/alibi_joint_qkv_attention.py |
69 |
6 |
0 |
20 |
7 |
83% |
| transformer_lens/model_bridge/generalized_components/alibi_utils.py |
19 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/model_bridge/generalized_components/altup_block.py |
58 |
15 |
0 |
22 |
7 |
62% |
| transformer_lens/model_bridge/generalized_components/attention.py |
370 |
29 |
0 |
170 |
34 |
87% |
| transformer_lens/model_bridge/generalized_components/audio_feature_extractor.py |
16 |
7 |
0 |
4 |
0 |
45% |
| transformer_lens/model_bridge/generalized_components/base.py |
221 |
33 |
0 |
120 |
15 |
82% |
| transformer_lens/model_bridge/generalized_components/block.py |
163 |
13 |
0 |
68 |
12 |
87% |
| transformer_lens/model_bridge/generalized_components/bloom_attention.py |
100 |
7 |
0 |
26 |
9 |
86% |
| transformer_lens/model_bridge/generalized_components/bloom_block.py |
42 |
23 |
0 |
16 |
2 |
36% |
| transformer_lens/model_bridge/generalized_components/bloom_mlp.py |
22 |
3 |
0 |
8 |
2 |
77% |
| transformer_lens/model_bridge/generalized_components/clip_vision_encoder.py |
43 |
17 |
0 |
14 |
1 |
51% |
| transformer_lens/model_bridge/generalized_components/codegen_attention.py |
85 |
17 |
0 |
18 |
6 |
72% |
| transformer_lens/model_bridge/generalized_components/conv1d.py |
20 |
12 |
0 |
4 |
0 |
33% |
| transformer_lens/model_bridge/generalized_components/conv_pos_embed.py |
16 |
7 |
0 |
4 |
0 |
45% |
| transformer_lens/model_bridge/generalized_components/depthwise_conv1d.py |
30 |
16 |
0 |
6 |
1 |
42% |
| transformer_lens/model_bridge/generalized_components/embedding.py |
44 |
12 |
0 |
16 |
7 |
65% |
| transformer_lens/model_bridge/generalized_components/gated_delta_net.py |
186 |
21 |
0 |
46 |
10 |
84% |
| transformer_lens/model_bridge/generalized_components/gated_mlp.py |
106 |
14 |
0 |
50 |
14 |
78% |
| transformer_lens/model_bridge/generalized_components/gated_rms_norm.py |
16 |
1 |
0 |
4 |
2 |
85% |
| transformer_lens/model_bridge/generalized_components/glm_moe_dsa_attention.py |
125 |
24 |
0 |
34 |
12 |
72% |
| transformer_lens/model_bridge/generalized_components/joint_gate_up_mlp.py |
73 |
7 |
0 |
20 |
5 |
83% |
| transformer_lens/model_bridge/generalized_components/joint_qkv_attention.py |
224 |
16 |
0 |
70 |
24 |
86% |
| transformer_lens/model_bridge/generalized_components/joint_qkv_position_embeddings_attention.py |
76 |
23 |
0 |
12 |
2 |
65% |
| transformer_lens/model_bridge/generalized_components/lfm2_gated_short_conv.py |
3 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/generalized_components/linear.py |
47 |
20 |
0 |
22 |
6 |
54% |
| transformer_lens/model_bridge/generalized_components/mla_attention.py |
178 |
54 |
0 |
50 |
4 |
64% |
| transformer_lens/model_bridge/generalized_components/mlp.py |
70 |
11 |
0 |
26 |
4 |
80% |
| transformer_lens/model_bridge/generalized_components/moe.py |
85 |
16 |
0 |
42 |
11 |
76% |
| transformer_lens/model_bridge/generalized_components/mpt_alibi_attention.py |
69 |
15 |
0 |
16 |
6 |
71% |
| transformer_lens/model_bridge/generalized_components/normalization.py |
100 |
7 |
0 |
42 |
6 |
91% |
| transformer_lens/model_bridge/generalized_components/opaque_block.py |
54 |
13 |
0 |
24 |
7 |
67% |
| transformer_lens/model_bridge/generalized_components/pos_embed.py |
30 |
2 |
0 |
12 |
2 |
90% |
| transformer_lens/model_bridge/generalized_components/position_embedding_hooks_mixin.py |
17 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/model_bridge/generalized_components/position_embeddings_attention.py |
300 |
48 |
0 |
118 |
18 |
81% |
| transformer_lens/model_bridge/generalized_components/qwen3_5_vision_encoder.py |
18 |
0 |
0 |
4 |
0 |
100% |
| transformer_lens/model_bridge/generalized_components/rms_normalization.py |
9 |
0 |
2 |
2 |
1 |
91% |
| transformer_lens/model_bridge/generalized_components/rotary_embedding.py |
52 |
14 |
0 |
24 |
9 |
64% |
| transformer_lens/model_bridge/generalized_components/siglip_vision_encoder.py |
43 |
17 |
0 |
14 |
1 |
51% |
| transformer_lens/model_bridge/generalized_components/ssm2_mixer.py |
127 |
3 |
0 |
30 |
7 |
94% |
| transformer_lens/model_bridge/generalized_components/ssm_block.py |
4 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/generalized_components/ssm_mixer.py |
154 |
6 |
0 |
48 |
7 |
93% |
| transformer_lens/model_bridge/generalized_components/ssm_protocol.py |
23 |
0 |
7 |
4 |
0 |
100% |
| transformer_lens/model_bridge/generalized_components/symbolic.py |
8 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/generalized_components/t5_block.py |
92 |
22 |
0 |
36 |
11 |
68% |
| transformer_lens/model_bridge/generalized_components/t5gemma2_decoder_block.py |
54 |
8 |
0 |
8 |
2 |
77% |
| transformer_lens/model_bridge/generalized_components/t5gemma2_merged_attention.py |
24 |
0 |
0 |
4 |
2 |
93% |
| transformer_lens/model_bridge/generalized_components/t5gemma_decoder_block.py |
64 |
38 |
0 |
8 |
0 |
36% |
| transformer_lens/model_bridge/generalized_components/unembedding.py |
59 |
17 |
0 |
20 |
7 |
65% |
| transformer_lens/model_bridge/generalized_components/vision_classifier_head.py |
11 |
1 |
0 |
2 |
1 |
85% |
| transformer_lens/model_bridge/generalized_components/vision_embeddings.py |
7 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/generalized_components/vision_projection.py |
18 |
2 |
0 |
4 |
2 |
82% |
| transformer_lens/model_bridge/get_params_util.py |
120 |
0 |
0 |
32 |
3 |
98% |
| transformer_lens/model_bridge/sources/__init__.py |
3 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/sources/_bridge_builder.py |
85 |
12 |
0 |
38 |
4 |
87% |
| transformer_lens/model_bridge/sources/native/__init__.py |
3 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/sources/native/init.py |
74 |
2 |
0 |
26 |
3 |
95% |
| transformer_lens/model_bridge/sources/native/model.py |
250 |
11 |
0 |
60 |
7 |
94% |
| transformer_lens/model_bridge/sources/transformers.py |
507 |
108 |
0 |
308 |
40 |
79% |
| transformer_lens/model_bridge/supported_architectures/__init__.py |
141 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/afmoe.py |
11 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/apertus.py |
63 |
3 |
0 |
6 |
2 |
93% |
| transformer_lens/model_bridge/supported_architectures/arcee.py |
16 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/ast.py |
30 |
1 |
0 |
6 |
3 |
89% |
| transformer_lens/model_bridge/supported_architectures/audio_flamingo3.py |
2 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/baichuan.py |
213 |
36 |
0 |
76 |
14 |
77% |
| transformer_lens/model_bridge/supported_architectures/bamba.py |
18 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/bart.py |
81 |
0 |
0 |
18 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/bd3lm.py |
64 |
13 |
0 |
14 |
3 |
77% |
| transformer_lens/model_bridge/supported_architectures/bert.py |
22 |
1 |
0 |
2 |
1 |
92% |
| transformer_lens/model_bridge/supported_architectures/bitnet.py |
19 |
0 |
0 |
2 |
1 |
95% |
| transformer_lens/model_bridge/supported_architectures/blenderbot.py |
5 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/bloom.py |
40 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/codegen.py |
33 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/cohere.py |
84 |
4 |
0 |
32 |
5 |
92% |
| transformer_lens/model_bridge/supported_architectures/deepseek_v2.py |
18 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/deepseek_v3.py |
16 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/deepseek_v4.py |
77 |
8 |
0 |
26 |
10 |
79% |
| transformer_lens/model_bridge/supported_architectures/dream.py |
64 |
18 |
0 |
16 |
3 |
61% |
| transformer_lens/model_bridge/supported_architectures/emu3.py |
9 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/ernie4_5.py |
8 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/ernie4_5_moe.py |
13 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/exaone.py |
27 |
12 |
0 |
4 |
0 |
48% |
| transformer_lens/model_bridge/supported_architectures/exaone4.py |
29 |
2 |
0 |
10 |
2 |
90% |
| transformer_lens/model_bridge/supported_architectures/falcon.py |
103 |
25 |
0 |
36 |
7 |
73% |
| transformer_lens/model_bridge/supported_architectures/falcon_h1.py |
29 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/falcon_mamba.py |
2 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/flex_olmo.py |
6 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/florence2.py |
13 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/gemma1.py |
14 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/gemma2.py |
19 |
2 |
0 |
4 |
2 |
83% |
| transformer_lens/model_bridge/supported_architectures/gemma3.py |
29 |
9 |
0 |
10 |
0 |
51% |
| transformer_lens/model_bridge/supported_architectures/gemma3_multimodal.py |
33 |
9 |
0 |
10 |
0 |
56% |
| transformer_lens/model_bridge/supported_architectures/gemma3n.py |
22 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/gemma4.py |
28 |
0 |
0 |
4 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/gemma4_text.py |
11 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/gidd.py |
56 |
18 |
0 |
8 |
1 |
64% |
| transformer_lens/model_bridge/supported_architectures/glm.py |
16 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/glm4.py |
5 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/glm4_moe.py |
12 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/glm4_moe_lite.py |
18 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/glm4v.py |
18 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/glm_asr.py |
2 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/glm_moe_dsa.py |
14 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/gpt2.py |
52 |
8 |
0 |
14 |
3 |
74% |
| transformer_lens/model_bridge/supported_architectures/gpt2_lm_head_custom.py |
10 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/gpt_bigcode.py |
62 |
0 |
0 |
10 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/gpt_oss.py |
23 |
0 |
0 |
8 |
1 |
97% |
| transformer_lens/model_bridge/supported_architectures/gptj.py |
16 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/granite.py |
24 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/granite_moe.py |
5 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/granite_moe_hybrid.py |
28 |
0 |
0 |
6 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/hrm_text.py |
58 |
23 |
0 |
32 |
2 |
48% |
| transformer_lens/model_bridge/supported_architectures/hubert.py |
40 |
0 |
0 |
4 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/hunyuan_v1_dense.py |
10 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/hyenadna.py |
25 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/idefics3.py |
14 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/internlm2.py |
168 |
40 |
0 |
52 |
8 |
69% |
| transformer_lens/model_bridge/supported_architectures/jais2.py |
8 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/jamba.py |
51 |
0 |
0 |
4 |
1 |
98% |
| transformer_lens/model_bridge/supported_architectures/jetmoe.py |
15 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/laguna.py |
22 |
2 |
0 |
4 |
2 |
85% |
| transformer_lens/model_bridge/supported_architectures/led.py |
11 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/lfm2.py |
35 |
0 |
0 |
18 |
5 |
91% |
| transformer_lens/model_bridge/supported_architectures/lfm2_moe.py |
28 |
0 |
0 |
12 |
2 |
95% |
| transformer_lens/model_bridge/supported_architectures/llada.py |
182 |
23 |
0 |
52 |
15 |
83% |
| transformer_lens/model_bridge/supported_architectures/llada2_moe.py |
46 |
1 |
0 |
10 |
2 |
95% |
| transformer_lens/model_bridge/supported_architectures/llama.py |
13 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/llama4.py |
31 |
3 |
0 |
8 |
3 |
85% |
| transformer_lens/model_bridge/supported_architectures/llama4_multimodal.py |
13 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/llava.py |
20 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/llava_next.py |
3 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/llava_onevision.py |
16 |
2 |
0 |
10 |
2 |
85% |
| transformer_lens/model_bridge/supported_architectures/longt5.py |
11 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/m2m100.py |
3 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/mamba.py |
24 |
1 |
0 |
2 |
1 |
92% |
| transformer_lens/model_bridge/supported_architectures/mamba2.py |
44 |
1 |
0 |
2 |
1 |
96% |
| transformer_lens/model_bridge/supported_architectures/marian.py |
2 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/mbart.py |
4 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/mingpt.py |
10 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/minimax_m2.py |
10 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/ministral3.py |
2 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/mistral.py |
9 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/mistral3.py |
7 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/mixtral.py |
10 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/modernbert_decoder.py |
18 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/mpt.py |
30 |
0 |
0 |
2 |
1 |
97% |
| transformer_lens/model_bridge/supported_architectures/music_flamingo.py |
2 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/nanochat.py |
18 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/nanogpt.py |
18 |
5 |
0 |
4 |
0 |
59% |
| transformer_lens/model_bridge/supported_architectures/native.py |
50 |
0 |
0 |
16 |
1 |
98% |
| transformer_lens/model_bridge/supported_architectures/neel_solu_old.py |
26 |
14 |
0 |
12 |
0 |
32% |
| transformer_lens/model_bridge/supported_architectures/nemotron.py |
16 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/nemotron_h.py |
32 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/neo.py |
34 |
0 |
0 |
4 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/neox.py |
53 |
0 |
0 |
6 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/olmo.py |
46 |
1 |
0 |
10 |
2 |
95% |
| transformer_lens/model_bridge/supported_architectures/olmo2.py |
14 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/olmo3.py |
3 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/olmo_hybrid.py |
22 |
1 |
0 |
2 |
1 |
92% |
| transformer_lens/model_bridge/supported_architectures/olmoe.py |
14 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/openai_gpt.py |
21 |
0 |
0 |
2 |
1 |
96% |
| transformer_lens/model_bridge/supported_architectures/openelm.py |
74 |
55 |
0 |
28 |
0 |
19% |
| transformer_lens/model_bridge/supported_architectures/opt.py |
26 |
0 |
0 |
6 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/ouro.py |
47 |
18 |
0 |
14 |
0 |
57% |
| transformer_lens/model_bridge/supported_architectures/pegasus.py |
3 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/phi.py |
19 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/phi3.py |
118 |
13 |
0 |
42 |
9 |
81% |
| transformer_lens/model_bridge/supported_architectures/phimoe.py |
41 |
3 |
0 |
18 |
2 |
88% |
| transformer_lens/model_bridge/supported_architectures/pretrain.py |
112 |
9 |
0 |
50 |
7 |
89% |
| transformer_lens/model_bridge/supported_architectures/qwen.py |
48 |
1 |
0 |
8 |
1 |
96% |
| transformer_lens/model_bridge/supported_architectures/qwen2.py |
14 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/qwen2_5_vl.py |
17 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/qwen2_audio.py |
12 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/qwen2_moe.py |
12 |
1 |
0 |
2 |
1 |
86% |
| transformer_lens/model_bridge/supported_architectures/qwen3.py |
39 |
0 |
0 |
6 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/qwen3_5.py |
21 |
0 |
0 |
4 |
1 |
96% |
| transformer_lens/model_bridge/supported_architectures/qwen3_5_moe.py |
30 |
0 |
0 |
4 |
1 |
97% |
| transformer_lens/model_bridge/supported_architectures/qwen3_5_multimodal.py |
19 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/qwen3_moe.py |
11 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/qwen3_next.py |
12 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/qwen3_vl.py |
21 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/qwen3_vl_moe.py |
6 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/raven.py |
53 |
22 |
0 |
10 |
0 |
49% |
| transformer_lens/model_bridge/supported_architectures/recurrent_gemma.py |
21 |
0 |
0 |
4 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/rwkv.py |
32 |
0 |
0 |
4 |
2 |
94% |
| transformer_lens/model_bridge/supported_architectures/rwkv7.py |
60 |
25 |
0 |
10 |
0 |
50% |
| transformer_lens/model_bridge/supported_architectures/seed_oss.py |
7 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/smollm3.py |
21 |
0 |
0 |
4 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/stablelm.py |
26 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/starcoder2.py |
15 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/switch_transformers.py |
20 |
1 |
0 |
0 |
0 |
95% |
| transformer_lens/model_bridge/supported_architectures/t5.py |
21 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/t5gemma.py |
36 |
14 |
0 |
12 |
0 |
46% |
| transformer_lens/model_bridge/supported_architectures/t5gemma2.py |
48 |
20 |
0 |
18 |
0 |
42% |
| transformer_lens/model_bridge/supported_architectures/vaultgemma.py |
13 |
0 |
0 |
4 |
1 |
94% |
| transformer_lens/model_bridge/supported_architectures/vit.py |
38 |
0 |
0 |
8 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/xglm.py |
15 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/youtu.py |
4 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/supported_architectures/zamba2.py |
29 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/model_bridge/types.py |
11 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/patching.py |
142 |
36 |
34 |
14 |
6 |
73% |
| transformer_lens/pretrained/__init__.py |
0 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/pretrained/weight_conversions/__init__.py |
28 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/pretrained/weight_conversions/apertus.py |
70 |
7 |
0 |
14 |
4 |
87% |
| transformer_lens/pretrained/weight_conversions/bert.py |
37 |
33 |
0 |
2 |
0 |
10% |
| transformer_lens/pretrained/weight_conversions/bloom.py |
40 |
36 |
0 |
2 |
0 |
10% |
| transformer_lens/pretrained/weight_conversions/coder.py |
43 |
38 |
0 |
2 |
0 |
11% |
| transformer_lens/pretrained/weight_conversions/gemma.py |
54 |
2 |
0 |
14 |
3 |
93% |
| transformer_lens/pretrained/weight_conversions/gpt2.py |
43 |
1 |
0 |
4 |
1 |
96% |
| transformer_lens/pretrained/weight_conversions/gptj.py |
36 |
31 |
0 |
2 |
0 |
13% |
| transformer_lens/pretrained/weight_conversions/hubert.py |
69 |
11 |
0 |
36 |
12 |
78% |
| transformer_lens/pretrained/weight_conversions/llama.py |
45 |
39 |
0 |
8 |
0 |
11% |
| transformer_lens/pretrained/weight_conversions/mingpt.py |
40 |
36 |
0 |
2 |
0 |
10% |
| transformer_lens/pretrained/weight_conversions/mistral.py |
36 |
31 |
0 |
2 |
0 |
13% |
| transformer_lens/pretrained/weight_conversions/mixtral.py |
37 |
32 |
0 |
4 |
0 |
12% |
| transformer_lens/pretrained/weight_conversions/nanogpt.py |
62 |
57 |
0 |
12 |
0 |
7% |
| transformer_lens/pretrained/weight_conversions/neel_solu_old.py |
17 |
14 |
0 |
12 |
0 |
10% |
| transformer_lens/pretrained/weight_conversions/neo.py |
40 |
0 |
0 |
4 |
1 |
98% |
| transformer_lens/pretrained/weight_conversions/neox.py |
34 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/pretrained/weight_conversions/olmo.py |
36 |
31 |
0 |
2 |
0 |
13% |
| transformer_lens/pretrained/weight_conversions/olmo2.py |
40 |
34 |
0 |
2 |
0 |
14% |
| transformer_lens/pretrained/weight_conversions/olmo3.py |
42 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/pretrained/weight_conversions/olmoe.py |
43 |
38 |
0 |
4 |
0 |
11% |
| transformer_lens/pretrained/weight_conversions/openai.py |
52 |
4 |
0 |
10 |
2 |
90% |
| transformer_lens/pretrained/weight_conversions/opt.py |
40 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/pretrained/weight_conversions/phi.py |
41 |
37 |
0 |
2 |
0 |
9% |
| transformer_lens/pretrained/weight_conversions/phi3.py |
41 |
35 |
0 |
2 |
0 |
14% |
| transformer_lens/pretrained/weight_conversions/qwen.py |
38 |
33 |
0 |
2 |
0 |
12% |
| transformer_lens/pretrained/weight_conversions/qwen2.py |
41 |
36 |
0 |
2 |
0 |
12% |
| transformer_lens/pretrained/weight_conversions/qwen3.py |
43 |
37 |
0 |
4 |
0 |
13% |
| transformer_lens/pretrained/weight_conversions/t5.py |
33 |
29 |
0 |
4 |
0 |
11% |
| transformer_lens/supported_models.py |
5 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/tools/__init__.py |
2 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/tools/analysis/__init__.py |
4 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/tools/analysis/direct_logit_attribution.py |
52 |
0 |
0 |
22 |
2 |
97% |
| transformer_lens/tools/analysis/direct_path_patching.py |
60 |
1 |
0 |
10 |
1 |
97% |
| transformer_lens/tools/analysis/jacobian_lens.py |
522 |
26 |
0 |
206 |
18 |
93% |
| transformer_lens/tools/model_registry/__init__.py |
6 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/tools/model_registry/alias_drift.py |
61 |
26 |
0 |
20 |
2 |
48% |
| transformer_lens/tools/model_registry/api.py |
142 |
43 |
0 |
52 |
9 |
62% |
| transformer_lens/tools/model_registry/discover_architectures.py |
100 |
100 |
0 |
30 |
0 |
0% |
| transformer_lens/tools/model_registry/exceptions.py |
36 |
18 |
0 |
10 |
1 |
41% |
| transformer_lens/tools/model_registry/generate_report.py |
113 |
102 |
0 |
34 |
1 |
8% |
| transformer_lens/tools/model_registry/hf_scraper.py |
351 |
351 |
6 |
142 |
0 |
0% |
| transformer_lens/tools/model_registry/registry_io.py |
103 |
7 |
0 |
36 |
4 |
89% |
| transformer_lens/tools/model_registry/relevancy.py |
40 |
0 |
0 |
18 |
0 |
100% |
| transformer_lens/tools/model_registry/schemas.py |
128 |
2 |
0 |
16 |
4 |
96% |
| transformer_lens/tools/model_registry/validate.py |
238 |
80 |
0 |
140 |
37 |
58% |
| transformer_lens/tools/model_registry/verification.py |
48 |
0 |
0 |
8 |
1 |
98% |
| transformer_lens/tools/model_registry/verify_models.py |
670 |
533 |
0 |
326 |
8 |
18% |
| transformer_lens/train.py |
75 |
45 |
0 |
32 |
0 |
28% |
| transformer_lens/utilities/__init__.py |
18 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/utilities/activation_functions.py |
46 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/utilities/addmm.py |
10 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/utilities/aliases.py |
59 |
8 |
0 |
36 |
2 |
87% |
| transformer_lens/utilities/architectures.py |
57 |
21 |
0 |
22 |
2 |
56% |
| transformer_lens/utilities/attention.py |
15 |
2 |
0 |
4 |
2 |
79% |
| transformer_lens/utilities/attribute_utils.py |
11 |
0 |
0 |
4 |
0 |
100% |
| transformer_lens/utilities/bridge_components.py |
44 |
5 |
0 |
24 |
6 |
84% |
| transformer_lens/utilities/cache.py |
6 |
0 |
0 |
2 |
0 |
100% |
| transformer_lens/utilities/components_utils.py |
29 |
3 |
0 |
18 |
3 |
87% |
| transformer_lens/utilities/defaults_utils.py |
34 |
0 |
0 |
10 |
0 |
100% |
| transformer_lens/utilities/devices.py |
63 |
0 |
5 |
34 |
3 |
97% |
| transformer_lens/utilities/exploratory_utils.py |
48 |
4 |
0 |
16 |
5 |
86% |
| transformer_lens/utilities/gpu_utils.py |
5 |
1 |
0 |
0 |
0 |
80% |
| transformer_lens/utilities/hf_utils.py |
123 |
35 |
0 |
44 |
6 |
67% |
| transformer_lens/utilities/initialization_utils.py |
41 |
0 |
0 |
8 |
0 |
100% |
| transformer_lens/utilities/library_utils.py |
4 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/utilities/lm_utils.py |
22 |
5 |
0 |
6 |
1 |
71% |
| transformer_lens/utilities/logits_utils.py |
56 |
1 |
0 |
20 |
1 |
97% |
| transformer_lens/utilities/matrix.py |
23 |
1 |
0 |
6 |
1 |
93% |
| transformer_lens/utilities/multi_gpu.py |
139 |
15 |
2 |
64 |
3 |
90% |
| transformer_lens/utilities/slice.py |
46 |
2 |
0 |
18 |
2 |
94% |
| transformer_lens/utilities/tensors.py |
87 |
34 |
0 |
42 |
1 |
54% |
| transformer_lens/utilities/tokenize_utils.py |
107 |
4 |
0 |
40 |
7 |
93% |
| transformer_lens/utilities/tracr.py |
60 |
4 |
0 |
18 |
5 |
88% |
| transformer_lens/utilities/typed_module_list.py |
4 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/utils.py |
4 |
0 |
0 |
0 |
0 |
100% |
| transformer_lens/weight_processing.py |
842 |
176 |
46 |
410 |
95 |
75% |