© 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. 8-RES_POST_32K-OAI
    11. 8477
    Prev
    Next
    INDEX
    Explanations

    tokens related to technical or formatting symbols, particularly in a structured data context

    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
    +0.17
    0.8%
    373
    +0.12
    0.6%
    109
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    647
    +0.17
    0.01
    497
    +0.12
    0.01
    207
    +0.09
    0.01
    Negative Logits
     REPL
    -0.83
     Authors
    -0.76
     Shepard
    -0.75
     Rebirth
    -0.74
     Incre
    -0.73
     Viol
    -0.72
     Abuse
    -0.71
     Transit
    -0.71
     Sakuya
    -0.70
     SOLD
    -0.69
    POSITIVE LOGITS
    dn
    1.15
    ngth
    1.13
    jri
    1.08
    phe
    1.06
    milo
    1.04
    dx
    1.01
    yip
    1.01
    bh
    0.99
    daq
    0.98
    ikh
    0.98
    Activations Density 0.037%

    No Known Activations