© 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. 30763
    Prev
    Next
    INDEX
    Explanations

    actions that express emotional release or relief

    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
    669
    +0.13
    0.6%
    534
    +0.12
    0.6%
    373
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    282
    +0.13
    0.06
    59
    +0.12
    0.03
    608
    +0.09
    0.05
    Negative Logits
    ibrary
    -0.73
     mortg
    -0.71
     municip
    -0.68
     lawy
    -0.66
     contrace
    -0.65
    luaj
    -0.63
    indal
    -0.63
     differe
    -0.63
     helicop
    -0.63
     ingred
    -0.62
    POSITIVE LOGITS
     a
    0.87
     another
    0.86
     an
    0.82
     Seconds
    0.79
    pause
    0.78
    amaz
    0.75
     grin
    0.74
     hesitation
    0.73
    Pause
    0.70
     glances
    0.70
    Activations Density 0.298%

    No Known Activations