© 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. 4182
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "token" and related terms within a programming context

    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
    288
    +0.17
    0.8%
    373
    +0.11
    0.5%
    739
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.17
    0.02
    473
    +0.11
    0.01
    326
    +0.11
    0.01
    Negative Logits
     Schr
    -0.71
    esome
    -0.68
    sonian
    -0.66
    ingly
    -0.66
    orld
    -0.65
     Sonic
    -0.65
    aukee
    -0.64
     Cald
    -0.63
    cens
    -0.62
    lla
    -0.62
    POSITIVE LOGITS
    omach
    0.91
     holders
    0.87
    holders
    0.84
    ozo
    0.83
    istically
    0.81
    izer
    0.80
     holder
    0.79
    istic
    0.78
     tokens
    0.78
    holder
    0.77
    Activations Density 0.054%

    No Known Activations