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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 6149
    Prev
    Next
    INDEX
    Explanations

    numbered lists of items

    oai_token-act-pair · gpt-3.5-turbo

    lists or enumerations of reasons or things

    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
    ±
    -0.75
    erity
    -0.70
    roth
    -0.67
    Lago
    -0.65
    rag
    -0.64
    grim
    -0.63
     Cros
    -0.62
    Ń·
    -0.62
    raf
    -0.61
    bor
    -0.61
    POSITIVE LOGITS
     categories
    0.74
     finalists
    0.72
     viz
    0.71
     contenders
    0.69
     storylines
    0.66
     acron
    0.66
     namely
    0.66
     personalities
    0.64
     explanations
    0.64
     theories
    0.64
    Activations Density 0.813%

    No Known Activations