© 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. 32281
    Prev
    Next
    INDEX
    Explanations

    specific instances of notable names and entities within a context

    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
    21
    +0.11
    0.5%
    373
    +0.10
    0.5%
    295
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    192
    +0.11
    0.04
    107
    +0.10
    0.05
    711
    +0.10
    0.03
    Negative Logits
    cffff
    -0.79
    utics
    -0.77
    reddits
    -0.74
    norm
    -0.72
    pse
    -0.71
    umi
    -0.70
    */(
    -0.70
    recy
    -0.68
    itton
    -0.66
    ignty
    -0.66
    POSITIVE LOGITS
    -
    0.93
    _
    0.70
     -
    0.70
    âĢij
    0.69
    +
    0.61
     +
    0.57
    _{
    0.55
    ÙĴ
    0.55
     slee
    0.55
     Span
    0.54
    Activations Density 0.322%

    No Known Activations