© 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. 31748
    Prev
    Next
    INDEX
    Explanations

    phrasal constructs with varying degrees of intensity or adequacy

    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
    94
    +0.11
    0.5%
    373
    +0.11
    0.5%
    9
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    94
    +0.11
    0.03
    7
    +0.11
    0.03
    438
    +0.09
    0.02
    Negative Logits
     incorrectly
    -0.70
    æ©Ł
    -0.68
     ado
    -0.66
     instead
    -0.66
     otherwise
    -0.65
     Reviews
    -0.65
     fewer
    -0.64
     errone
    -0.64
    Others
    -0.64
     wrongly
    -0.64
    POSITIVE LOGITS
    enough
    0.76
     ready
    0.71
    ready
    0.71
     sure
    0.70
     Enough
    0.70
    orious
    0.68
     finished
    0.67
     perfected
    0.66
    stock
    0.66
    uring
    0.65
    Activations Density 0.182%

    No Known Activations