© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 4824
    Prev
    Next
    INDEX
    Explanations

    references to legal statutes and regulations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    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
    373
    +0.11
    0.5%
    417
    +0.11
    0.5%
    612
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.04
    548
    +0.11
    0.02
    404
    +0.11
    0.03
    Negative Logits
    ded
    -0.93
    mers
    -0.87
    matter
    -0.86
    wig
    -0.85
    ted
    -0.85
    ming
    -0.85
    bur
    -0.80
    mos
    -0.79
    umin
    -0.78
    pher
    -0.78
    POSITIVE LOGITS
    entimes
    0.73
    ENSE
    0.72
     Revised
    0.70
     Division
    0.69
     Weekly
    0.69
     Utilities
    0.68
    ĵ
    0.68
     Code
    0.67
     Reference
    0.65
     offense
    0.64
    Activations Density 0.167%

    No Known Activations