Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 12911
    Prev
    Next
    INDEX
    Explanations

    references to physical actions involving the use of fists

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "fist" and its variations in different contexts

    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
    Marginal
    -0.89
    Trend
    -0.82
     Agency
    -0.82
    Sleep
    -0.77
     Comfort
    -0.74
    acent
    -0.71
    actionDate
    -0.71
    Smart
    -0.71
     Living
    -0.69
    Leaks
    -0.69
    POSITIVE LOGITS
     clen
    1.12
     clenched
    1.10
     fist
    1.05
     fists
    0.97
    fights
    0.90
     salute
    0.89
    pak
    0.83
    rag
    0.81
     smack
    0.78
    ulas
    0.77
    Activations Density 0.008%

    No Known Activations