© 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. 4-RES-JB
    11. 1007
    Prev
    Next
    INDEX
    Explanations

    mentions of individuals or figures related to Hall of Fame

    oai_token-act-pair · gpt-3.5-turbo

    references to the Hall of Fame

    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
     Cosponsors
    -0.85
    uyomi
    -0.83
     contrace
    -0.77
    nets
    -0.71
    irie
    -0.71
    inational
    -0.69
    perm
    -0.68
     Uriel
    -0.67
     guiActiveUn
    -0.67
    ģ«
    -0.67
    POSITIVE LOGITS
     Fame
    0.71
     elders
    0.68
    athe
    0.66
     integrity
    0.65
     glory
    0.63
     Presidents
    0.60
     Gry
    0.60
    erential
    0.58
     horrors
    0.57
     Giants
    0.56
    Activations Density 0.103%

    No Known Activations