© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 29117
    Prev
    Next
    INDEX
    Explanations

    references to scientific research or analysis methods

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    269
    +0.12
    0.5%
    440
    +0.09
    0.4%
    109
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    236
    +0.12
    0.12
    498
    +0.09
    0.11
    408
    +0.08
    0.14
    Negative Logits
    soType
    -0.91
     Tradable
    -0.84
    gencies
    -0.84
    iscons
    -0.78
    iors
    -0.78
    iciary
    -0.76
    acia
    -0.76
    anooga
    -0.75
    bernatorial
    -0.75
     grievance
    -0.75
    POSITIVE LOGITS
     researchers
    1.61
     scientists
    1.52
     physicists
    1.34
     biologists
    1.31
    Researchers
    1.28
     astronomers
    1.23
     molecular
    1.22
     Researchers
    1.19
    Scientists
    1.17
     computational
    1.16
    Activations Density 1.933%

    No Known Activations