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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 24267
    Prev
    Next
    INDEX
    Explanations

    words related to physical impacts or collisions

    oai_token-act-pair · gpt-3.5-turbo

    references related to the concept of "bumping" or encountering obstacles

    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
    eval
    -0.78
    ufact
    -0.76
    encia
    -0.74
    RIP
    -0.73
    ANCE
    -0.73
     Mechdragon
    -0.67
    soDeliveryDate
    -0.65
    natureconservancy
    -0.64
     Ecology
    -0.64
     ministry
    -0.63
    POSITIVE LOGITS
    kins
    0.92
     bump
    0.89
    sticks
    0.87
    kin
    0.87
     bumps
    0.85
    olean
    0.83
    stick
    0.82
    balls
    0.82
    saw
    0.81
    alos
    0.79
    Activations Density 0.051%

    No Known Activations