© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 26211
    Prev
    Next
    INDEX
    Explanations

    expressions of excessiveness or comparison

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    192
    +0.13
    0.6%
    57
    +0.11
    0.5%
    216
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.13
    0.13
    540
    +0.11
    0.11
    190
    +0.09
    0.09
    Negative Logits
    culosis
    -0.87
    ngth
    -0.84
    velt
    -0.79
    assador
    -0.74
    hea
    -0.74
    ratulations
    -0.73
    rection
    -0.72
    loads
    -0.72
     Commissioners
    -0.70
    iership
    -0.70
    POSITIVE LOGITS
     abstract
    1.36
     simplistic
    1.30
     disparate
    1.29
     rudimentary
    1.17
     neb
    1.16
     vague
    1.14
     generic
    1.14
     esoteric
    1.13
     diffuse
    1.12
     interchangeable
    1.12
    Activations Density 1.107%

    No Known Activations