© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 10-RES_POST_32K-OAI
    11. 29729
    Prev
    Next
    INDEX
    Explanations

    instances of the word "by" and related phrases indicating attribution or authorship

    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
    493
    +0.11
    0.5%
    657
    +0.11
    0.5%
    476
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    476
    +0.11
    0.01
    232
    +0.11
    0.03
    22
    +0.11
    0.02
    Negative Logits
     disabling
    -0.67
     prematurely
    -0.66
     disembark
    -0.63
    CRIP
    -0.63
    IBLE
    -0.63
    theless
    -0.63
    nown
    -0.62
     facult
    -0.62
    ulnerable
    -0.61
    mble
    -0.61
    POSITIVE LOGITS
     Stats
    0.80
     Proposition
    0.70
     Reps
    0.67
    lip
    0.65
     129
    0.65
    cheon
    0.64
    ondo
    0.64
     Acts
    0.61
     131
    0.60
     Sketch
    0.60
    Activations Density 0.165%

    No Known Activations