© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 6001
    Prev
    Next
    INDEX
    Explanations

    mentions of substances and regulations related to controlled substances

    oai_token-act-pair · gpt-3.5-turbo

    references to controlled substances and their regulation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Honour
    -0.85
    åĭ
    -0.75
     Shards
    -0.73
    é¾įåĸļ士
    -0.70
     honoured
    -0.69
    Na
    -0.69
     Thieves
    -0.69
     Kinnikuman
    -0.68
     Archdemon
    -0.67
    Sac
    -0.66
    POSITIVE LOGITS
     controlled
    1.20
     contro
    0.96
    controlled
    0.95
    poons
    0.85
     Controlled
    0.83
    ording
    0.79
     runaway
    0.79
    creen
    0.78
    ependent
    0.77
     substances
    0.76
    Activations Density 0.018%

    No Known Activations