© 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. 5122
    Prev
    Next
    INDEX
    Explanations

    references to creative works and titles

    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
    728
    +0.10
    0.4%
    734
    +0.09
    0.4%
    521
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.10
    0.13
    232
    +0.09
    0.13
    284
    +0.09
    0.09
    Negative Logits
    cliffe
    -0.77
    minecraft
    -0.77
    schild
    -0.75
    inki
    -0.68
     hurd
    -0.60
     fare
    -0.60
     shove
    -0.59
    dylib
    -0.59
    iland
    -0.59
    assetsadobe
    -0.58
    POSITIVE LOGITS
    ioxide
    0.81
    ciating
    0.72
    rette
    0.71
    fil
    0.70
    bourg
    0.67
    ©¶æ
    0.66
    Enh
    0.63
    onential
    0.62
    quez
    0.62
    Neg
    0.60
    Activations Density 3.516%

    No Known Activations