© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 28008
    Prev
    Next
    INDEX
    Explanations

    references to patches and related activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    243
    +0.11
    0.5%
    113
    +0.10
    0.4%
    288
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.02
    466
    +0.10
    0.02
    651
    +0.10
    0.02
    Negative Logits
     Flavoring
    -0.71
     Galile
    -0.69
    ccording
    -0.69
    emouth
    -0.68
     carniv
    -0.66
     contrad
    -0.65
    VERTISEMENT
    -0.65
     obser
    -0.65
     hemor
    -0.62
    cius
    -0.61
    POSITIVE LOGITS
    work
    1.31
    notes
    1.02
    ions
    0.98
    oul
    0.96
    ioned
    0.94
    ogue
    0.85
    hern
    0.84
    ting
    0.83
    ing
    0.81
    intosh
    0.81
    Activations Density 0.090%

    No Known Activations