Transluce/features_explain_llama3.1_8b_simulator
8B
•
Updated
•
48
Training Language Models To Explain Their Own Computations
Models and datasets for "Training Language Models To Explain Their Own Computations"
-
-
Transluce/act_patch_llama_3.1_8b_counterfact
Viewer • Updated • 126k • 40 -
Transluce/input_ablation_qwen3_8b_mmlu_hint
Viewer • Updated • 14k • 29 • 1 -
Transluce/input_ablation_llama_3.1_8b_instruct_mmlu_hint
Viewer • Updated • 14k • 28 -
Transluce/act_patch_qwen3_8b_counterfact
Viewer • Updated • 135k • 31 -
Transluce/features_explain_llama3.1_8b_llama3.1_8b
Text Generation • 8B • Updated • 26 -
Transluce/features_explain_llama3.1_8b_llama3.1_8b_instruct
Text Generation • 8B • Updated • 24 -
Transluce/features_explain_llama3.1_8b_llama3_8b
Text Generation • 8B • Updated • 11 -
Transluce/act_patch_llama3.1_8b_llama3.1_8b
Text Generation • Updated • 10 -
Transluce/act_patch_qwen3_8b_qwen3_8b
Text Generation • Updated • 11 -
Transluce/input_ablation_llama3.1_8b_instruct_llama3.1_8b_instruct
Text Generation • 8B • Updated • 19 -
Transluce/input_ablation_qwen3_8b_qwen3_8b_hint
Text Generation • 8B • Updated • 15