© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 8652
    Prev
    Next
    INDEX
    Explanations

    quotation marks, indicating dialogue or citations

    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
    314
    +0.16
    0.7%
    55
    +0.11
    0.5%
    426
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    426
    +0.16
    0.02
    314
    +0.11
    0.00
    442
    +0.10
    0.02
    Negative Logits
     caut
    -0.66
    aser
    -0.64
     abduct
    -0.64
     predomin
    -0.63
    Ͻ
    -0.62
     arsen
    -0.61
     dignity
    -0.61
    asers
    -0.59
     caric
    -0.59
     lapt
    -0.58
    POSITIVE LOGITS
    ãģ¨
    0.77
    undo
    0.74
    alias
    0.73
     />
    0.73
    ][/
    0.72
    uns
    0.71
     ++
    0.69
    isSpecialOrderable
    0.69
    tails
    0.68
     %%
    0.67
    Activations Density 0.065%

    No Known Activations