© 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. 10-RES_POST_32K-OAI
    11. 27656
    Prev
    Next
    INDEX
    Explanations

    references to content related to writing and publishing

    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
    208
    +0.14
    0.6%
    301
    +0.13
    0.6%
    471
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    208
    +0.14
    0.01
    301
    +0.13
    0.01
    398
    +0.09
    0.01
    Negative Logits
    CLASSIFIED
    -0.74
    otropic
    -0.67
    rolet
    -0.63
     carcin
    -0.60
     satisf
    -0.60
     gangs
    -0.58
    Completed
    -0.58
    ocene
    -0.58
    agin
    -0.58
     Allaah
    -0.57
    POSITIVE LOGITS
    river
    0.73
    sit
    0.72
    mad
    0.71
    ĸļ
    0.66
    UME
    0.65
    ume
    0.64
    ayer
    0.63
    endor
    0.62
    encia
    0.62
    asma
    0.62
    Activations Density 0.094%

    No Known Activations