© 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. 7922
    Prev
    Next
    INDEX
    Explanations

    dialogue or interactions concerning conflict or tension

    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
    373
    +0.13
    0.6%
    502
    +0.11
    0.5%
    456
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.13
    0.07
    640
    +0.11
    0.07
    107
    +0.11
    0.06
    Negative Logits
    cius
    -0.72
     improbable
    -0.65
     indo
    -0.65
     flourish
    -0.65
     scen
    -0.64
     rul
    -0.63
     glyph
    -0.61
     droid
    -0.60
     untold
    -0.58
     unthinkable
    -0.58
    POSITIVE LOGITS
    uting
    0.79
    ãĥ¼ãĥ³
    0.77
    ibly
    0.72
    ãĤī
    0.70
    ãĤ®
    0.68
    κ
    0.68
    ãĥ¼ãĤ¯
    0.66
    vik
    0.65
    lished
    0.65
    itness
    0.65
    Activations Density 0.266%

    No Known Activations