© 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. 32678
    Prev
    Next
    INDEX
    Explanations

    words related to emotional expressions and human interactions

    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
    312
    +0.12
    0.5%
    388
    +0.10
    0.5%
    544
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.12
    -0.01
    284
    +0.10
    0.05
    406
    +0.10
    0.06
    Negative Logits
    rences
    -0.70
    soType
    -0.68
    heastern
    -0.67
     Genie
    -0.67
     Siren
    -0.65
    anmar
    -0.64
    heast
    -0.64
     veter
    -0.63
     Sakuya
    -0.62
     pse
    -0.60
    POSITIVE LOGITS
    ruck
    0.78
    eed
    0.70
    andise
    0.69
    acks
    0.68
    addle
    0.68
    acked
    0.65
    oled
    0.65
    load
    0.64
    atts
    0.64
    un
    0.63
    Activations Density 1.308%

    No Known Activations