© 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. 4791
    Prev
    Next
    INDEX
    Explanations

    overall user ratings and reviews for media 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
    39
    +0.13
    0.6%
    550
    +0.11
    0.5%
    191
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    39
    +0.13
    0.04
    739
    +0.11
    0.01
    222
    +0.10
    0.03
    Negative Logits
    nown
    -0.81
    İĭ
    -0.67
     spaces
    -0.66
     disgu
    -0.65
    esi
    -0.64
     refuge
    -0.63
     manif
    -0.62
    figure
    -0.62
     Wonderland
    -0.62
     Shell
    -0.61
    POSITIVE LOGITS
     rating
    1.00
     ratings
    0.95
     Ratings
    0.93
    rating
    0.91
     Rated
    0.88
    votes
    0.87
     rated
    0.85
     biased
    0.82
     Rating
    0.81
    review
    0.81
    Activations Density 1.214%

    No Known Activations