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

    Neuronpedia

    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. 7564
    Prev
    Next
    INDEX
    Explanations

    the last name "Adkins" at varying levels of activation

    oai_token-act-pair · gpt-3.5-turbo

    the surname "kins" in various 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
    ntil
    -0.71
    ccording
    -0.71
    senal
    -0.69
     Seym
    -0.67
    ħĭ
    -0.67
    undo
    -0.65
    ascal
    -0.63
    astical
    -0.61
    PsyNetMessage
    -0.61
    iners
    -0.61
    POSITIVE LOGITS
    kins
    1.14
    hips
    1.11
    hip
    1.09
    haw
    1.06
    creen
    1.04
    kie
    0.99
    burg
    0.94
    cence
    0.94
    sey
    0.93
    kin
    0.91
    Activations Density 0.009%

    No Known Activations