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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 7201
    Prev
    Next
    INDEX
    Explanations

    days of the week

    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
    430
    +0.18
    0.9%
    105
    +0.11
    0.5%
    333
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    430
    +0.18
    0.05
    105
    +0.11
    0.04
    384
    +0.10
    0.03
    Negative Logits
    geries
    -0.75
    FontSize
    -0.61
    Translation
    -0.60
    Exit
    -0.60
     Bucket
    -0.60
    EStreamFrame
    -0.59
    gery
    -0.58
     Defendants
    -0.57
     Alchemy
    -0.57
    aic
    -0.57
    POSITIVE LOGITS
     undertook
    1.06
     withdrew
    1.01
     reacted
    1.01
     launched
    1.01
     embarked
    1.00
     underwent
    0.97
     teamed
    0.97
     reaff
    0.97
     apologized
    0.96
     announced
    0.94
    Activations Density 0.097%

    No Known Activations