© 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 EvalsExportsAPI 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. 10922
    Prev
    Next
    INDEX
    Explanations

    text containing special characters and symbols like âĢĶ and âĢĵ

    oai_token-act-pair · gpt-3.5-turbo

    punctuation and special characters

    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
     thous
    -0.80
    istries
    -0.72
     Lama
    -0.70
    Ãį
    -0.67
     nodd
    -0.65
    gerald
    -0.65
     destro
    -0.64
    cedes
    -0.62
     challeng
    -0.62
    emort
    -0.62
    POSITIVE LOGITS
    utical
    0.80
    USS
    0.74
    =-=-=-=-
    0.69
    OIL
    0.67
    uv
    0.67
    frac
    0.66
     uh
    0.63
    veh
    0.62
    andel
    0.62
    hover
    0.61
    Activations Density 0.314%

    No Known Activations