© 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. 4973
    Prev
    Next
    INDEX
    Explanations

    mentions of various types of content

    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
    496
    +0.13
    0.6%
    439
    +0.12
    0.6%
    298
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    439
    +0.13
    0.04
    298
    +0.12
    0.03
    495
    +0.10
    0.02
    Negative Logits
     Siem
    -0.71
     Sons
    -0.64
    vals
    -0.63
     moss
    -0.63
     damn
    -0.61
    west
    -0.59
    JD
    -0.59
     Jazz
    -0.57
     Cot
    -0.57
     Galile
    -0.57
    POSITIVE LOGITS
    edly
    1.33
     creators
    0.90
    ions
    0.76
    eval
    0.75
    ed
    0.74
    ivist
    0.74
    ibilities
    0.72
    Ö¼
    0.71
     creator
    0.69
    ioned
    0.69
    Activations Density 0.090%

    No Known Activations