© 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. 10-RES_POST_32K-OAI
    11. 31000
    Prev
    Next
    INDEX
    Explanations

    references to biting and bites, particularly in the context of animals or pests

    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
    366
    +0.14
    0.6%
    433
    +0.09
    0.4%
    50
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    366
    +0.14
    0.02
    750
    +0.09
    0.02
    526
    +0.09
    -0.00
    Negative Logits
    oppable
    -0.81
    ourced
    -0.74
     Princ
    -0.71
    ALSE
    -0.66
     Fargo
    -0.64
     Abbey
    -0.63
     Mathemat
    -0.63
     Bland
    -0.62
    YC
    -0.62
    ivari
    -0.61
    POSITIVE LOGITS
     bite
    1.05
    bite
    0.97
     gum
    0.90
     bites
    0.85
     bitten
    0.84
     bait
    0.77
    paste
    0.75
     Bite
    0.73
    vine
    0.72
     wounds
    0.71
    Activations Density 0.154%

    No Known Activations