[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6QNMC2GJ7SR2WAQ3JCZ2TVZ":3,"skill-same:01M6QNMC2GJ7SR2WAQ3JCZ2TVZ":96},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":59,"commitSha":60,"skillDirectory":61,"forksCount":14,"repositoryHtmlUrl":62,"repositoryHomepageUrl":63,"ownerHtmlUrl":64,"ownerBlogUrl":65,"ownerDescription":66,"updatedAt":67,"install":68,"resources":81},"01M6QNMC2GJ7SR2WAQ3JCZ2TVZ","transformer-lens-interpretability",null,"Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering model algorithms, studying attention patterns, or performing activation patching experiments.","Tashima-Tarsh\u002FDisha6.6","Tashima-Tarsh","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F269906498?v=4","01M6PBH98X03XERH0HFT4PFB1P","NITISH KUMAR","01M6PBH9DA0M6KM3Z12P848F0K",0,"MIT","https:\u002F\u002Fgithub.com\u002FTashima-Tarsh\u002FDisha6.6\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fqc-transformer-lens\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000021","资料检索与研究","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"transformer lens interpretability","transformer lens","Interpretability","Transformer","lens","for","research","inspect","circuit","TransformerLens","circuit analysis","attention","algorithms","guidance","use","model","patching","reverse-engineering","engineering","experiments","2026-11-10 10:12:30",false,"---\nname: transformer-lens-interpretability\ndescription: Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering model algorithms, studying attention patterns, or performing activation patching experiments.\nversion: 1.0.0\nauthor: Orchestra Research\nlicense: MIT\ntags: [Mechanistic Interpretability, TransformerLens, Activation Patching, Circuit Analysis]\ndependencies: [transformer-lens>=2.0.0, torch>=2.0.0]\ncategory: \"ai-research\"\n---\n\n# TransformerLens: Mechanistic Interpretability for Transformers\n\nTransformerLens is the de facto standard library for mechanistic interpretability research on GPT-style language models. Created by Neel Nanda and maintained by Bryce Meyer, it provides clean interfaces to inspect and manipulate model internals via HookPoints on every activation.\n\n**GitHub**: [TransformerLensOrg\u002FTransformerLens](https:\u002F\u002Fgithub.com\u002FTransformerLensOrg\u002FTransformerLens) (2,900+ stars)\n\n## When to Use TransformerLens\n\n**Use TransformerLens when you need to:**\n- Reverse-engineer algorithms learned during training\n- Perform activation patching \u002F causal tracing experiments\n- Study attention patterns and information flow\n- Analyze circuits (e.g., induction heads, IOI circuit)\n- Cache and inspect intermediate activations\n- Apply direct logit attribution\n\n**Consider alternatives when:**\n- You need to work with non-transformer architectures → Use **nnsight** or **pyvene**\n- You want to train\u002Fanalyze Sparse Autoencoders → Use **SAELens**\n- You need remote execution on massive models → Use **nnsight** with NDIF\n- You want higher-level causal intervention abstractions → Use **pyvene**\n\n## Installation\n\n```bash\npip install transformer-lens\n```\n\nFor development version:\n```bash\npip install git+https:\u002F\u002Fgithub.com\u002FTransformerLensOrg\u002FTransformerLens\n```\n\n## Core Concepts\n\n### HookedTransformer\n\nThe main class that wraps transformer models with HookPoints on every activation:\n\n```python\nfrom transformer_lens import HookedTransformer\n\n# Load a model\nmodel = HookedTransformer.from_pretrained(\"gpt2-small\")\n\n# For gated models (LLaMA, Mistral)\nimport os\nos.environ[\"HF_TOKEN\"] = \"your_token\"\nmodel = HookedTransformer.from_pretrained(\"meta-llama\u002FLlama-2-7b-hf\")\n```\n\n### Supported Models (50+)\n\n| Family | Models |\n|--------|--------|\n| GPT-2 | gpt2, gpt2-medium, gpt2-large, gpt2-xl |\n| LLaMA | llama-7b, llama-13b, llama-2-7b, llama-2-13b |\n| EleutherAI | pythia-70m to pythia-12b, gpt-neo, gpt-j-6b |\n| Mistral | mistral-7b, mixtral-8x7b |\n| Others | phi, qwen, opt, gemma |\n\n### Activation Caching\n\nRun the model and cache all intermediate activations:\n\n```python\n# Get all activations\ntokens = model.to_tokens(\"The Eiffel Tower is in\")\nlogits, cache = model.run_with_cache(tokens)\n\n# Access specific activations\nresidual = cache[\"resid_post\", 5]  # Layer 5 residual stream\nattn_pattern = cache[\"pattern\", 3]  # Layer 3 attention pattern\nmlp_out = cache[\"mlp_out\", 7]  # Layer 7 MLP output\n\n# Filter which activations to cache (saves memory)\nlogits, cache = model.run_with_cache(\n    tokens,\n    names_filter=lambda name: \"resid_post\" in name\n)\n```\n\n### ActivationCache Keys\n\n| Key Pattern | Shape | Description |\n|-------------|-------|-------------|\n| `resid_pre, layer` | [batch, pos, d_model] | Residual before attention |\n| `resid_mid, layer` | [batch, pos, d_model] | Residual after attention |\n| `resid_post, layer` | [batch, pos, d_model] | Residual after MLP |\n| `attn_out, layer` | [batch, pos, d_model] | Attention output |\n| `mlp_out, layer` | [batch, pos, d_model] | MLP output |\n| `pattern, layer` | [batch, head, q_pos, k_pos] | Attention pattern (post-softmax) |\n| `q, layer` | [batch, pos, head, d_head] | Query vectors |\n| `k, layer` | [batch, pos, head, d_head] | Key vectors |\n| `v, layer` | [batch, pos, head, d_head] | Value vectors |\n\n## Workflow 1: Activation Patching (Causal Tracing)\n\nIdentify which activations causally affect model output by patching clean activations into corrupted runs.\n\n### Step-by-Step\n\n```python\nfrom transformer_lens import HookedTransformer, patching\nimport torch\n\nmodel = HookedTransformer.from_pretrained(\"gpt2-small\")\n\n# 1. Define clean and corrupted prompts\nclean_prompt = \"The Eiffel Tower is in the city of\"\ncorrupted_prompt = \"The Colosseum is in the city of\"\n\nclean_tokens = model.to_tokens(clean_prompt)\ncorrupted_tokens = model.to_tokens(corrupted_prompt)\n\n# 2. Get clean activations\n_, clean_cache = model.run_with_cache(clean_tokens)\n\n# 3. Define metric (e.g., logit difference)\nparis_token = model.to_single_token(\" Paris\")\nrome_token = model.to_single_token(\" Rome\")\n\ndef metric(logits):\n    return logits[0, -1, paris_token] - logits[0, -1, rome_token]\n\n# 4. Patch each position and layer\nresults = torch.zeros(model.cfg.n_layers, clean_tokens.shape[1])\n\nfor layer in range(model.cfg.n_layers):\n    for pos in range(clean_tokens.shape[1]):\n        def patch_hook(activation, hook):\n            activation[0, pos] = clean_cache[hook.name][0, pos]\n            return activation\n\n        patched_logits = model.run_with_hooks(\n            corrupted_tokens,\n            fwd_hooks=[(f\"blocks.{layer}.hook_resid_post\", patch_hook)]\n        )\n        results[layer, pos] = metric(patched_logits)\n\n# 5. Visualize results (layer x position heatmap)\n```\n\n### Checklist\n- [ ] Define clean and corrupted inputs that differ minimally\n- [ ] Choose metric that captures behavior difference\n- [ ] Cache clean activations\n- [ ] Systematically patch each (layer, position) combination\n- [ ] Visualize results as heatmap\n- [ ] Identify causal hotspots\n\n## Workflow 2: Circuit Analysis (Indirect Object Identification)\n\nReplicate the IOI circuit discovery from \"Interpretability in the Wild\".\n\n### Step-by-Step\n\n```python\nfrom transformer_lens import HookedTransformer\nimport torch\n\nmodel = HookedTransformer.from_pretrained(\"gpt2-small\")\n\n# IOI task: \"When John and Mary went to the store, Mary gave a bottle to\"\n# Model should predict \"John\" (indirect object)\n\nprompt = \"When John and Mary went to the store, Mary gave a bottle to\"\ntokens = model.to_tokens(prompt)\n\n# 1. Get baseline logits\nlogits, cache = model.run_with_cache(tokens)\n\njohn_token = model.to_single_token(\" John\")\nmary_token = model.to_single_token(\" Mary\")\n\n# 2. Compute logit difference (IO - S)\nlogit_diff = logits[0, -1, john_token] - logits[0, -1, mary_token]\nprint(f\"Logit difference: {logit_diff.item():.3f}\")\n\n# 3. Direct logit attribution by head\ndef get_head_contribution(layer, head):\n    # Project head output to logits\n    head_out = cache[\"z\", layer][0, :, head, :]  # [pos, d_head]\n    W_O = model.W_O[layer, head]  # [d_head, d_model]\n    W_U = model.W_U  # [d_model, vocab]\n\n    # Head contribution to logits at final position\n    contribution = head_out[-1] @ W_O @ W_U\n    return contribution[john_token] - contribution[mary_token]\n\n# 4. Map all heads\nhead_contributions = torch.zeros(model.cfg.n_layers, model.cfg.n_heads)\nfor layer in range(model.cfg.n_layers):\n    for head in range(model.cfg.n_heads):\n        head_contributions[layer, head] = get_head_contribution(layer, head)\n\n# 5. Identify top contributing heads (name movers, backup name movers)\n```\n\n### Checklist\n- [ ] Set up task with clear IO\u002FS tokens\n- [ ] Compute baseline logit difference\n- [ ] Decompose by attention head contributions\n- [ ] Identify key circuit components (name movers, S-inhibition, induction)\n- [ ] Validate with ablation experiments\n\n## Workflow 3: Induction Head Detection\n\nFind induction heads that implement [A][B]...[A] → [B] pattern.\n\n```python\nfrom transformer_lens import HookedTransformer\nimport torch\n\nmodel = HookedTransformer.from_pretrained(\"gpt2-small\")\n\n# Create repeated sequence: [A][B][A] should predict [B]\nrepeated_tokens = torch.tensor([[1000, 2000, 1000]])  # Arbitrary tokens\n\n_, cache = model.run_with_cache(repeated_tokens)\n\n# Induction heads attend from final [A] back to first [B]\n# Check attention from position 2 to position 1\ninduction_scores = torch.zeros(model.cfg.n_layers, model.cfg.n_heads)\n\nfor layer in range(model.cfg.n_layers):\n    pattern = cache[\"pattern\", layer][0]  # [head, q_pos, k_pos]\n    # Attention from pos 2 to pos 1\n    induction_scores[layer] = pattern[:, 2, 1]\n\n# Heads with high scores are induction heads\ntop_heads = torch.topk(induction_scores.flatten(), k=5)\n```\n\n## Common Issues & Solutions\n\n### Issue: Hooks persist after debugging\n```python\n# WRONG: Old hooks remain active\nmodel.run_with_hooks(tokens, fwd_hooks=[...])  # Debug, add new hooks\nmodel.run_with_hooks(tokens, fwd_hooks=[...])  # Old hooks still there!\n\n# RIGHT: Always reset hooks\nmodel.reset_hooks()\nmodel.run_with_hooks(tokens, fwd_hooks=[...])\n```\n\n### Issue: Tokenization gotchas\n```python\n# WRONG: Assuming consistent tokenization\nmodel.to_tokens(\"Tim\")  # Single token\nmodel.to_tokens(\"Neel\")  # Becomes \"Ne\" + \"el\" (two tokens!)\n\n# RIGHT: Check tokenization explicitly\ntokens = model.to_tokens(\"Neel\", prepend_bos=False)\nprint(model.to_str_tokens(tokens))  # ['Ne', 'el']\n```\n\n### Issue: LayerNorm ignored in analysis\n```python\n# WRONG: Ignoring LayerNorm\npre_activation = residual @ model.W_in[layer]\n\n# RIGHT: Include LayerNorm\nln_scale = model.blocks[layer].ln2.w\nln_out = model.blocks[layer].ln2(residual)\npre_activation = ln_out @ model.W_in[layer]\n```\n\n### Issue: Memory explosion with large models\n```python\n# Use selective caching\nlogits, cache = model.run_with_cache(\n    tokens,\n    names_filter=lambda n: \"resid_post\" in n or \"pattern\" in n,\n    device=\"cpu\"  # Cache on CPU\n)\n```\n\n## Key Classes Reference\n\n| Class | Purpose |\n|-------|---------|\n| `HookedTransformer` | Main model wrapper with hooks |\n| `ActivationCache` | Dictionary-like cache of activations |\n| `HookedTransformerConfig` | Model configuration |\n| `FactoredMatrix` | Efficient factored matrix operations |\n\n## Integration with SAELens\n\nTransformerLens integrates with SAELens for Sparse Autoencoder analysis:\n\n```python\nfrom transformer_lens import HookedTransformer\nfrom sae_lens import SAE\n\nmodel = HookedTransformer.from_pretrained(\"gpt2-small\")\nsae = SAE.from_pretrained(\"gpt2-small-res-jb\", \"blocks.8.hook_resid_pre\")\n\n# Run with SAE\ntokens = model.to_tokens(\"Hello world\")\n_, cache = model.run_with_cache(tokens)\nsae_acts = sae.encode(cache[\"resid_pre\", 8])\n```\n\n## Reference Documentation\n\nFor detailed API documentation, tutorials, and advanced usage, see the `references\u002F` folder:\n\n| File | Contents |\n|------|----------|\n| [references\u002FREADME.md](references\u002FREADME.md) | Overview and quick start guide |\n| [references\u002Fapi.md](references\u002Fapi.md) | Complete API reference for HookedTransformer, ActivationCache, HookPoints |\n| [references\u002Ftutorials.md](references\u002Ftutorials.md) | Step-by-step tutorials for activation patching, circuit analysis, logit lens |\n\n## External Resources\n\n### Tutorials\n- [Main Demo Notebook](https:\u002F\u002Ftransformerlensorg.github.io\u002FTransformerLens\u002Fgenerated\u002Fdemos\u002FMain_Demo.html)\n- [Activation Patching Demo](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FTransformerLensOrg\u002FTransformerLens\u002Fblob\u002Fmain\u002Fdemos\u002FActivation_Patching_in_TL_Demo.ipynb)\n- [ARENA Mech Interp Course](https:\u002F\u002Farena-foundation.github.io\u002FARENA\u002F) - 200+ hours of tutorials\n\n### Papers\n- [A Mathematical Framework for Transformer Circuits](https:\u002F\u002Ftransformer-circuits.pub\u002F2021\u002Fframework\u002Findex.html)\n- [In-context Learning and Induction Heads](https:\u002F\u002Ftransformer-circuits.pub\u002F2022\u002Fin-context-learning-and-induction-heads\u002Findex.html)\n- [Interpretability in the Wild (IOI)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.00593)\n\n### Official Documentation\n- [Official Docs](https:\u002F\u002Ftransformerlensorg.github.io\u002FTransformerLens\u002F)\n- [Model Properties Table](https:\u002F\u002Ftransformerlensorg.github.io\u002FTransformerLens\u002Fgenerated\u002Fmodel_properties_table.html)\n- [Neel Nanda's Glossary](https:\u002F\u002Fwww.neelnanda.io\u002Fmechanistic-interpretability\u002Fglossary)\n\n## Version Notes\n\n- **v2.0**: Removed HookedSAE (moved to SAELens)\n- **v3.0 (alpha)**: TransformerBridge for loading any nn.Module\n","# TransformerLens: Mechanistic Interpretability for Transformers\n\nTransformerLens is the de facto standard library for mechanistic interpretability research on GPT-style language models. Created by Neel Nanda and maintained by Bryce Meyer, it provides clean interfaces to inspect and manipulate model internals via HookPoints on every activation.\n\n**GitHub**: [TransformerLensOrg\u002FTransformerLens](https:\u002F\u002Fgithub.com\u002FTransformerLensOrg\u002FTransformerLens) (2,900+ stars)\n\n## When to Use TransformerLens\n\n**Use TransformerLens when you need to:**\n- Reverse-engineer algorithms learned during training\n- Perform activation patching \u002F causal tracing experiments\n- Study attention patterns and information flow\n- Analyze circuits (e.g., induction heads, IOI circuit)\n- Cache and inspect intermediate activations\n- Apply direct logit attribution\n\n**Consider alternatives when:**\n- You need to work with non-transformer architectures → Use **nnsight** or **pyvene**\n- You want to train\u002Fanalyze Sparse Autoencoders → Use **SAELens**\n- You need remote execution on massive models → Use **nnsight** with NDIF\n- You want higher-level causal intervention abstractions → Use **pyvene**\n\n## Installation\n\n```bash\npip install transformer-lens\n```\n\nFor development version:\n```bash\npip install git+https:\u002F\u002Fgithub.com\u002FTransformerLensOrg\u002FTransformerLens\n```\n\n## Core Concepts\n\n### HookedTransformer\n\nThe main class that wraps transformer models with HookPoints on every activation:\n\n```python\nfrom transformer_lens import HookedTransformer\n\n# Load a model\nmodel = HookedTransformer.from_pretrained(\"gpt2-small\")\n\n# For gated models (LLaMA, Mistral)\nimport os\nos.environ[\"HF_TOKEN\"] = \"your_token\"\nmodel = HookedTransformer.from_pretrained(\"meta-llama\u002FLlama-2-7b-hf\")\n```\n\n### Supported Models (50+)\n\n| Family | Models |\n|--------|--------|\n| GPT-2 | gpt2, gpt2-medium, gpt2-large, gpt2-xl |\n| LLaMA | llama-7b, llama-13b, llama-2-7b, llama-2-13b |\n| EleutherAI | pythia-70m to pythia-12b, gpt-neo, gpt-j-6b |\n| Mistral | mistral-7b, mixtral-8x7b |\n| Others | phi, qwen, opt, gemma |\n\n### Activation Caching\n\nRun the model and cache all intermediate activations:\n\n```python\n# Get all activations\ntokens = model.to_tokens(\"The Eiffel Tower is in\")\nlogits, cache = model.run_with_cache(tokens)\n\n# Access specific activations\nresidual = cache[\"resid_post\", 5]  # Layer 5 residual stream\nattn_pattern = cache[\"pattern\", 3]  # Layer 3 attention pattern\nmlp_out = cache[\"mlp_out\", 7]  # Layer 7 MLP output\n\n# Filter which activations to cache (saves memory)\nlogits, cache = model.run_with_cache(\n    tokens,\n    names_filter=lambda name: \"resid_post\" in name\n)\n```\n\n### ActivationCache Keys\n\n| Key Pattern | Shape | Description |\n|-------------|-------|-------------|\n| `resid_pre, layer` | [batch, pos, d_model] | Residual before attention |\n| `resid_mid, layer` | [batch, pos, d_model] | Residual after attention |\n| `resid_post, layer` | [batch, pos, d_model] | Residual after MLP |\n| `attn_out, layer` | [batch, pos, d_model] | Attention output |\n| `mlp_out, layer` | [batch, pos, d_model] | MLP output |\n| `pattern, layer` | [batch, head, q_pos, k_pos] | Attention pattern (post-softmax) |\n| `q, layer` | [batch, pos, head, d_head] | Query vectors |\n| `k, layer` | [batch, pos, head, d_head] | Key vectors |\n| `v, layer` | [batch, pos, head, d_head] | Value vectors |\n\n## Workflow 1: Activation Patching (Causal Tracing)\n\nIdentify which activations causally affect model output by patching clean activations into corrupted runs.\n\n### Step-by-Step\n\n```python\nfrom transformer_lens import HookedTransformer, patching\nimport torch\n\nmodel = HookedTransformer.from_pretrained(\"gpt2-small\")\n\n# 1. Define clean and corrupted prompts\nclean_prompt = \"The Eiffel Tower is in the city of\"\ncorrupted_prompt = \"The Colosseum is in the city of\"\n\nclean_tokens = model.to_tokens(clean_prompt)\ncorrupted_tokens = model.to_tokens(corrupted_prompt)\n\n# 2. Get clean activations\n_, clean_cache = model.run_with_cache(clean_tokens)\n\n# 3. Define metric (e.g., logit difference)\nparis_token = model.to_single_token(\" Paris\")\nrome_token = model.to_single_token(\" Rome\")\n\ndef metric(logits):\n    return logits[0, -1, paris_token] - logits[0, -1, rome_token]\n\n# 4. Patch each position and layer\nresults = torch.zeros(model.cfg.n_layers, clean_tokens.shape[1])\n\nfor layer in range(model.cfg.n_layers):\n    for pos in range(clean_tokens.shape[1]):\n        def patch_hook(activation, hook):\n            activation[0, pos] = clean_cache[hook.name][0, pos]\n            return activation\n\n        patched_logits = model.run_with_hooks(\n            corrupted_tokens,\n            fwd_hooks=[(f\"blocks.{layer}.hook_resid_post\", patch_hook)]\n        )\n        results[layer, pos] = metric(patched_logits)\n\n# 5. Visualize results (layer x position heatmap)\n```\n\n### Checklist\n- [ ] Define clean and corrupted inputs that differ minimally\n- [ ] Choose metric that captures behavior difference\n- [ ] Cache clean activations\n- [ ] Systematically patch each (layer, position) combination\n- [ ] Visualize results as heatmap\n- [ ] Identify causal hotspots\n\n## Workflow 2: Circuit Analysis (Indirect Object Identification)\n\nReplicate the IOI circuit discovery from \"Interpretability in the Wild\".\n\n### Step-by-Step\n\n```python\nfrom transformer_lens import HookedTransformer\nimport torch\n\nmodel = HookedTransformer.from_pretrained(\"gpt2-small\")\n\n# IOI task: \"When John and Mary went to the store, Mary gave a bottle to\"\n# Model should predict \"John\" (indirect object)\n\nprompt = \"When John and Mary went to the store, Mary gave a bottle to\"\ntokens = model.to_tokens(prompt)\n\n# 1. Get baseline logits\nlogits, cache = model.run_with_cache(tokens)\n\njohn_token = model.to_single_token(\" John\")\nmary_token = model.to_single_token(\" Mary\")\n\n# 2. Compute logit difference (IO - S)\nlogit_diff = logits[0, -1, john_token] - logits[0, -1, mary_token]\nprint(f\"Logit difference: {logit_diff.item():.3f}\")\n\n# 3. Direct logit attribution by head\ndef get_head_contribution(layer, head):\n    # Project head output to logits\n    head_out = cache[\"z\", layer][0, :, head, :]  # [pos, d_head]\n    W_O = model.W_O[layer, head]  # [d_head, d_model]\n    W_U = model.W_U  # [d_model, vocab]\n\n    # Head contribution to logits at final position\n    contribution = head_out[-1] @ W_O @ W_U\n    return contribution[john_token] - contribution[mary_token]\n\n# 4. Map all heads\nhead_contributions = torch.zeros(model.cfg.n_layers, model.cfg.n_heads)\nfor layer in range(model.cfg.n_layers):\n    for head in range(model.cfg.n_heads):\n        head_contributions[layer, head] = get_head_contribution(layer, head)\n\n# 5. Identify top contributing heads (name movers, backup name movers)\n```\n\n### Checklist\n- [ ] Set up task with clear IO\u002FS tokens\n- [ ] Compute baseline logit difference\n- [ ] Decompose by attention head contributions\n- [ ] Identify key circuit components (name movers, S-inhibition, induction)\n- [ ] Validate with ablation experiments\n\n## Workflow 3: Induction Head Detection\n\nFind induction heads that implement [A][B]...[A] → [B] pattern.\n\n```python\nfrom transformer_lens import HookedTransformer\nimport torch\n\nmodel = HookedTransformer.from_pretrained(\"gpt2-small\")\n\n# Create repeated sequence: [A][B][A] should predict [B]\nrepeated_tokens = torch.tensor([[1000, 2000, 1000]])  # Arbitrary tokens\n\n_, cache = model.run_with_cache(repeated_tokens)\n\n# Induction heads attend from final [A] back to first [B]\n# Check attention from position 2 to position 1\ninduction_scores = torch.zeros(model.cfg.n_layers, model.cfg.n_heads)\n\nfor layer in range(model.cfg.n_layers):\n    pattern = cache[\"pattern\", layer][0]  # [head, q_pos, k_pos]\n    # Attention from pos 2 to pos 1\n    induction_scores[layer] = pattern[:, 2, 1]\n\n# Heads with high scores are induction heads\ntop_heads = torch.topk(induction_scores.flatten(), k=5)\n```\n\n## Common Issues & Solutions\n\n### Issue: Hooks persist after debugging\n```python\n# WRONG: Old hooks remain active\nmodel.run_with_hooks(tokens, fwd_hooks=[...])  # Debug, add new hooks\nmodel.run_with_hooks(tokens, fwd_hooks=[...])  # Old hooks still there!\n\n# RIGHT: Always reset hooks\nmodel.reset_hooks()\nmodel.run_with_hooks(tokens, fwd_hooks=[...])\n```\n\n### Issue: Tokenization gotchas\n```python\n# WRONG: Assuming consistent tokenization\nmodel.to_tokens(\"Tim\")  # Single token\nmodel.to_tokens(\"Neel\")  # Becomes \"Ne\" + \"el\" (two tokens!)\n\n# RIGHT: Check tokenization explicitly\ntokens = model.to_tokens(\"Neel\", prepend_bos=False)\nprint(model.to_str_tokens(tokens))  # ['Ne', 'el']\n```\n\n### Issue: LayerNorm ignored in analysis\n```python\n# WRONG: Ignoring LayerNorm\npre_activation = residual @ model.W_in[layer]\n\n# RIGHT: Include LayerNorm\nln_scale = model.blocks[layer].ln2.w\nln_out = model.blocks[layer].ln2(residual)\npre_activation = ln_out @ model.W_in[layer]\n```\n\n### Issue: Memory explosion with large models\n```python\n# Use selective caching\nlogits, cache = model.run_with_cache(\n    tokens,\n    names_filter=lambda n: \"resid_post\" in n or \"pattern\" in n,\n    device=\"cpu\"  # Cache on CPU\n)\n```\n\n## Key Classes Reference\n\n| Class | Purpose |\n|-------|---------|\n| `HookedTransformer` | Main model wrapper with hooks |\n| `ActivationCache` | Dictionary-like cache of activations |\n| `HookedTransformerConfig` | Model configuration |\n| `FactoredMatrix` | Efficient factored matrix operations |\n\n## Integration with SAELens\n\nTransformerLens integrates with SAELens for Sparse Autoencoder analysis:\n\n```python\nfrom transformer_lens import HookedTransformer\nfrom sae_lens import SAE\n\nmodel = HookedTransformer.from_pretrained(\"gpt2-small\")\nsae = SAE.from_pretrained(\"gpt2-small-res-jb\", \"blocks.8.hook_resid_pre\")\n\n# Run with SAE\ntokens = model.to_tokens(\"Hello world\")\n_, cache = model.run_with_cache(tokens)\nsae_acts = sae.encode(cache[\"resid_pre\", 8])\n```\n\n## Reference Documentation\n\nFor detailed API documentation, tutorials, and advanced usage, see the `references\u002F` folder:\n\n| File | Contents |\n|------|----------|\n| [references\u002FREADME.md](references\u002FREADME.md) | Overview and quick start guide |\n| [references\u002Fapi.md](references\u002Fapi.md) | Complete API reference for HookedTransformer, ActivationCache, HookPoints |\n| [references\u002Ftutorials.md](references\u002Ftutorials.md) | Step-by-step tutorials for activation patching, circuit analysis, logit lens |\n\n## External Resources\n\n### Tutorials\n- [Main Demo Notebook](https:\u002F\u002Ftransformerlensorg.github.io\u002FTransformerLens\u002Fgenerated\u002Fdemos\u002FMain_Demo.html)\n- [Activation Patching Demo](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FTransformerLensOrg\u002FTransformerLens\u002Fblob\u002Fmain\u002Fdemos\u002FActivation_Patching_in_TL_Demo.ipynb)\n- [ARENA Mech Interp Course](https:\u002F\u002Farena-foundation.github.io\u002FARENA\u002F) - 200+ hours of tutorials\n\n### Papers\n- [A Mathematical Framework for Transformer Circuits](https:\u002F\u002Ftransformer-circuits.pub\u002F2021\u002Fframework\u002Findex.html)\n- [In-context Learning and Induction Heads](https:\u002F\u002Ftransformer-circuits.pub\u002F2022\u002Fin-context-learning-and-induction-heads\u002Findex.html)\n- [Interpretability in the Wild (IOI)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.00593)\n\n### Official Documentation\n- [Official Docs](https:\u002F\u002Ftransformerlensorg.github.io\u002FTransformerLens\u002F)\n- [Model Properties Table](https:\u002F\u002Ftransformerlensorg.github.io\u002FTransformerLens\u002Fgenerated\u002Fmodel_properties_table.html)\n- [Neel Nanda's Glossary](https:\u002F\u002Fwww.neelnanda.io\u002Fmechanistic-interpretability\u002Fglossary)\n\n## Version Notes\n\n- **v2.0**: Removed HookedSAE (moved to SAELens)\n- **v3.0 (alpha)**: TransformerBridge for loading any nn.Module\n",{"name":5,"tags":49,"author":53,"license":15,"version":54,"category":55,"description":7,"dependencies":56},[50,33,51,52],"Mechanistic Interpretability","Activation Patching","Circuit Analysis","Orchestra Research","1.0.0","ai-research",[57,58],"transformer-lens>=2.0.0","torch>=2.0.0","main","0da4485785ee599388900be50db1635489d53056",".agents\u002Fskills\u002Fqc-transformer-lens","https:\u002F\u002Fgithub.com\u002FTashima-Tarsh\u002FDisha6.6","https:\u002F\u002Fdisha6.6.thenitishkr.in\u002F","https:\u002F\u002Fgithub.com\u002FTashima-Tarsh","www.thenitishkr","thenitishkr | AI • Cyber-security • Research\r\n","2026-11-08 03:33:25",{"source":8,"skillName":5,"commands":69,"prompt":79,"zipAvailable":80},[70,73,76],{"tool":71,"command":72},"npx","npx skills add Tashima-Tarsh\u002FDisha6.6 --skill transformer-lens-interpretability",{"tool":74,"command":75},"bunx","bunx skills add Tashima-Tarsh\u002FDisha6.6 --skill transformer-lens-interpretability",{"tool":77,"command":78},"pnpm","pnpm dlx skills add Tashima-Tarsh\u002FDisha6.6 --skill transformer-lens-interpretability","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FTashima-Tarsh\u002FDisha6.6\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fqc-transformer-lens\u002FSKILL.md",true,[82,88,92],{"path":83,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":86,"sourceHtmlUrl":87},"references\u002Fapi.md","REFERENCE","参考文档",8349,"https:\u002F\u002Fgithub.com\u002FTashima-Tarsh\u002FDisha6.6\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fqc-transformer-lens\u002Freferences\u002Fapi.md",{"path":89,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":90,"sourceHtmlUrl":91},"references\u002FREADME.md",1643,"https:\u002F\u002Fgithub.com\u002FTashima-Tarsh\u002FDisha6.6\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fqc-transformer-lens\u002Freferences\u002FREADME.md",{"path":93,"resourceType":84,"resourceTypeDesc":85,"sizeBytes":94,"sourceHtmlUrl":95},"references\u002Ftutorials.md",10142,"https:\u002F\u002Fgithub.com\u002FTashima-Tarsh\u002FDisha6.6\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fqc-transformer-lens\u002Freferences\u002Ftutorials.md",[97,126,148,175,200,224],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":106,"publishedAt":44},"01M6QRD07D4TGNDK6VJ4EHF8BV","youtube-full","Use when the user needs YouTube transcripts, video search, channel browsing, playlist extraction, or content monitoring. Trigger phrases: 'get the transcript for', 'search YouTube for', 'what are the latest videos on', 'list this playlist', 'monitor this channel', or any request involving a YouTube URL, video ID, or @handle. Do NOT use for downloading video or audio files, YouTube engagement data (likes, comments), or private\u002Fage-restricted videos.","https:\u002F\u002Fgithub.com\u002FTashima-Tarsh\u002FDisha6.6\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fyoutube-full\u002FSKILL.md",[103],{"categoryNo":104,"categoryName":105,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体",[107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,29,123,124,125],"youtube full","youtube","full","transcript","search","monitoring","data","Video","Audio","list","files","do","get","age","trigger","URL","browsing","request","youtube transcripts",{"skillNo":127,"name":128,"nameZh":6,"description":129,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":130,"categories":131,"keywords":132,"publishedAt":44},"01M6QRCJ3112VYPA97YG7KGMC2","x-twitter-growth","X\u002FTwitter growth engine for building audience, crafting viral content, and analyzing engagement. Use when the user wants to grow on X\u002FTwitter, write tweets or threads, analyze their X profile, research competitors on X, plan a posting strategy, or optimize engagement. Complements social-content (generic multi-platform) with X-specific depth: algorithm mechanics, thread engineering, reply strategy, profile optimization, and competitive intelligence via web search.","https:\u002F\u002Fgithub.com\u002FTashima-Tarsh\u002FDisha6.6\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fx-twitter-growth\u002FSKILL.md",[],[133,134,135,30,111,136,42,137,138,139,140,141,142,143,38,144,145,146,147,29],"twitter","x twitter growth","growth","content","plan","web search","competitors","analyze","engine","write","strategy","platform","grow","posting","competitive",{"skillNo":149,"name":150,"nameZh":6,"description":151,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":152,"categories":153,"keywords":157,"publishedAt":44},"01M6QRBN3EHQNJY42RERP7A149","write-a-skill","Create new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, build, or author a new skill.","https:\u002F\u002Fgithub.com\u002FTashima-Tarsh\u002FDisha6.6\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fwrite-a-skill\u002FSKILL.md",[154],{"categoryNo":155,"categoryName":156,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[158,142,159,160,161,162,163,164,38,165,166,59,167,168,169,170,171,172,173,174],"write a skill","github","create","https","productivity","matt pocock","direct","example","build","new skill","structure","new agent","resources","with","tree","progressive disclosure","author",{"skillNo":176,"name":177,"nameZh":6,"description":178,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":179,"categories":180,"keywords":181,"publishedAt":44},"01M6QRB6G8P33PNGAF9Y69EJ9T","wrap-up","Close out a launched Claude Managed Agent — recap every primitive the founder now owns, regenerate the single-file overview page, and suggest the next 1-2 upgrades. Use when the user says \"wrap up\", \"close this out\", \"what do I own now\", \"give me the summary\", \"recap the agent\", or when the orchestrator routes phase=wrap-up. primitives_inventory.py tables everything owned (agent, environment, session, memory, outcome, deployment); overview_page.py regenerates a self-contained .\u002Fmy-agent\u002Fagent-overview.html; upgrade_suggester.py ranks the next moves from recorded deferrals plus standing hardening steps. Companion to run-without-you; the last stop before phase=done.","https:\u002F\u002Fgithub.com\u002FTashima-Tarsh\u002FDisha6.6\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fwrap-up\u002FSKILL.md",[],[182,183,184,185,186,187,188,189,190,38,191,118,192,193,194,195,196,197,198,199],"wrap up","up","wrap","gemini","deployment","run","file","hardening","closeout","stop","upgrade","founder","close","companion","you","tables","directions","phase",{"skillNo":201,"name":202,"nameZh":6,"description":203,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":204,"categories":205,"keywords":207,"publishedAt":44},"01M6QRA5AYC1RBFZ39GTWGS3N8","workflow-builder","Design and write deterministic multi-agent workflow scripts (.js files in .claude\u002Fworkflows\u002F) for Claude Code's Workflow tool. Use when a user wants to build, create, author, scaffold, or run a custom Claude Code workflow, orchestrate sub-agents (fan-out, pipeline, loop, judge-panel), or automate a repeatable multi-step task across fresh-context agents.","https:\u002F\u002Fgithub.com\u002FTashima-Tarsh\u002FDisha6.6\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fworkflow-builder\u002FSKILL.md",[206],{"categoryNo":155,"categoryName":156,"taxonomyType":21,"taxonomyTypeDesc":22},[208,209,210,211,212,160,187,213,214,215,216,217,142,218,117,219,220,221,222,223],"workflow","workflow builder","builder","design","workflows","code","orchestrate","pipeline","loop","multi-agent","context","across","fresh","scripts","judge","claude code workflow",{"skillNo":225,"name":226,"nameZh":6,"description":227,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":228,"categories":229,"keywords":230,"publishedAt":44},"01M6QR7QMCZESKQA0ERW1E657M","weekly-review","Use when someone wants to run a weekly review, close open loops, audit stalled projects and commitments, get their system back to trusted, restart a lapsed review habit, or says \"\u002Fcs:weekly-review\". Walks David Allen's three-phase loop — GET CLEAR, GET CURRENT, GET CREATIVE — with deterministic scripts that inventory open loops, gate the checklist with named gaps, and score commitment health 0-100.","https:\u002F\u002Fgithub.com\u002FTashima-Tarsh\u002FDisha6.6\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fweekly-review\u002FSKILL.md",[],[231,232,233,234,235,236,237,238,239,216,240,241,242,243,244,245,119,246,247,199],"review","weekly review","weekly","deterministic scripts","reflect","feeds","inventory","open","team","projects","project-management","scanner","conversation","back","project","sprint","cadence"]