© 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. 4669
    Prev
    Next
    INDEX
    Explanations

    quotes or speech within the text

    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
    734
    +0.09
    0.4%
    170
    +0.09
    0.4%
    627
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    107
    +0.09
    0.04
    6
    +0.09
    0.04
    246
    +0.09
    0.04
    Negative Logits
    FORE
    -0.77
    cffffcc
    -0.75
    MIT
    -0.74
    esson
    -0.72
    aceae
    -0.72
    Ms
    -0.71
     Wikimedia
    -0.70
    schild
    -0.69
    ibur
    -0.69
     legislatures
    -0.68
    POSITIVE LOGITS
    Coach
    1.09
     preseason
    1.06
     rookies
    1.06
     Coach
    1.03
     teammate
    1.02
     coach
    1.00
     coaching
    0.98
     Rookie
    0.98
     Defensive
    0.97
     teammates
    0.97
    Activations Density 0.188%

    No Known Activations