© 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 EvalsExportsAPI 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. 6337
    Prev
    Next
    INDEX
    Explanations

    references to seasons and installments of television shows and series

    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
    525
    +0.12
    0.6%
    464
    +0.11
    0.5%
    695
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.12
    0.09
    16
    +0.11
    0.04
    393
    +0.10
    0.07
    Negative Logits
     nods
    -0.71
     scratches
    -0.70
    auna
    -0.69
     touches
    -0.69
    wolves
    -0.67
     inputs
    -0.67
    abilia
    -0.66
    eton
    -0.66
    anchez
    -0.64
    forts
    -0.64
    POSITIVE LOGITS
     series
    1.04
    series
    1.04
     trilogy
    0.97
     TAMADRA
    0.85
    Series
    0.81
     Series
    0.76
     season
    0.74
    secut
    0.73
     saga
    0.69
     cycle
    0.67
    Activations Density 0.677%

    No Known Activations