© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 8-RES_POST_32K-OAI
    11. 5821
    Prev
    Next
    INDEX
    Explanations

    underscores in programming contexts, often associated with naming conventions or formatting

    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
    +0.18
    0.8%
    636
    +0.12
    0.6%
    640
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.18
    0.04
    442
    +0.12
    0.04
    107
    +0.11
    0.04
    Negative Logits
     Reloaded
    -0.82
     dexter
    -0.73
     Canaver
    -0.73
     Suite
    -0.73
     Machina
    -0.69
     Plaza
    -0.68
     Liberation
    -0.67
     Scare
    -0.66
     Qiao
    -0.66
     vans
    -0.66
    POSITIVE LOGITS
    existing
    1.22
    foreign
    1.09
    selection
    1.07
    local
    1.06
    paren
    1.03
    primary
    1.03
    focus
    0.97
    internet
    0.96
    shared
    0.94
    secondary
    0.93
    Activations Density 0.111%

    No Known Activations