© 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. 26279
    Prev
    Next
    INDEX
    Explanations

    assertions about proof or evidence of various claims

    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
    449
    +0.10
    0.5%
    12
    +0.10
    0.5%
    673
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.10
    0.05
    540
    +0.10
    0.04
    7
    +0.09
    0.04
    Negative Logits
    mentioned
    -0.75
    zza
    -0.71
    idia
    -0.70
    £ı
    -0.70
     foreground
    -0.69
    ALT
    -0.68
    ))))
    -0.68
    letal
    -0.66
    chedel
    -0.66
    ãĥ¼ãĥ«
    -0.65
    POSITIVE LOGITS
     invaluable
    1.06
     prophetic
    1.02
     fruitful
    1.02
     resilient
    0.95
     unreliable
    0.92
     fertile
    0.92
     fatal
    0.91
     costly
    0.89
     disastrous
    0.89
     profitable
    0.89
    Activations Density 0.514%

    No Known Activations