© 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. 11276
    Prev
    Next
    INDEX
    Explanations

    proper nouns, particularly names

    oai_token-act-pair · gpt-3.5-turbo

    the name "Steve" and its variations in different 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
    hazard
    -0.80
    ãĥīãĥ©
    -0.80
    merce
    -0.71
    ospace
    -0.68
    ä
    -0.68
    sembly
    -0.68
    urate
    -0.67
    flies
    -0.66
     [|
    -0.65
    naire
    -0.65
    POSITIVE LOGITS
    rers
    0.96
    rer
    0.95
    Ste
    0.91
    vey
    0.85
    rence
    0.85
    eding
    0.78
    fan
    0.77
    lla
    0.77
    aling
    0.75
    ved
    0.75
    Activations Density 0.009%

    No Known Activations