© 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. 7116
    Prev
    Next
    INDEX
    Explanations

    mentions of social media handles or users

    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
    314
    +0.23
    1.1%
    373
    +0.14
    0.6%
    627
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    627
    +0.23
    0.04
    445
    +0.14
    0.04
    396
    +0.10
    0.03
    Negative Logits
     prolifer
    -0.68
     concentrate
    -0.68
     strokes
    -0.67
     favor
    -0.67
     assum
    -0.67
     circulation
    -0.66
     nasal
    -0.66
     compulsion
    -0.66
     slips
    -0.66
     triangles
    -0.66
    POSITIVE LOGITS
    realDonaldTrump
    1.59
    #$
    1.41
    gmail
    1.21
    abc
    1.02
    GOP
    1.02
    @@@@@@@@
    1.02
    lat
    1.01
    example
    0.99
    TPS
    0.99
    INC
    0.97
    Activations Density 0.062%

    No Known Activations