© 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. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 20982
    Prev
    Next
    INDEX
    Explanations

    mentions of specific names or entities related to various events or topics

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    314
    +0.18
    0.9%
    496
    +0.12
    0.6%
    160
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.18
    0.03
    314
    +0.12
    0.01
    17
    +0.10
    0.02
    Negative Logits
     transported
    -0.75
     cones
    -0.75
     poisoning
    -0.74
     allowance
    -0.73
     uncertain
    -0.73
     appreci
    -0.72
     meaningful
    -0.72
     appropri
    -0.71
     values
    -0.71
     equivalents
    -0.70
    POSITIVE LOGITS
    âĢ¦]
    1.57
    ...]
    1.55
    !]
    1.31
    Pg
    1.24
    Laughs
    1.20
    email
    1.20
    img
    1.17
    ?]
    1.16
    AUT
    1.15
    UPDATE
    1.14
    Activations Density 0.023%

    No Known Activations