© 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. 8-RES-JB
    11. 20209
    Prev
    Next
    INDEX
    Explanations

    human body parts

    oai_token-act-pair · gpt-3.5-turbo

    references to people, organizations, or specific identifiers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    fig
    -0.73
    vine
    -0.71
    lance
    -0.70
    bestos
    -0.70
     Fet
    -0.69
    paren
    -0.68
    ribune
    -0.63
    glomer
    -0.63
    TPS
    -0.62
    utherland
    -0.62
    POSITIVE LOGITS
     suffers
    0.87
     prefers
    0.84
     underwent
    0.83
     enjoys
    0.82
     engaged
    0.81
    's
    0.80
     agrees
    0.80
     knew
    0.80
     undertook
    0.79
     possesses
    0.79
    Activations Density 0.824%

    No Known Activations