© 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. 7431
    Prev
    Next
    INDEX
    Explanations

    prepositions followed by a verb indicating direction or movement

    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
    218
    +0.11
    0.5%
    147
    +0.10
    0.4%
    419
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.11
    0.06
    362
    +0.10
    0.04
    526
    +0.09
    0.00
    Negative Logits
    primary
    -0.70
    SELECT
    -0.67
     SELECT
    -0.65
    BALL
    -0.65
    cmd
    -0.61
    estate
    -0.61
     exerted
    -0.60
     pioneered
    -0.60
    caster
    -0.60
    interstitial
    -0.60
    POSITIVE LOGITS
     obscurity
    1.24
     landfill
    1.03
     ashes
    1.00
     flames
    0.94
     unpop
    0.92
    livion
    0.92
     rubble
    0.92
     limbo
    0.87
    emetery
    0.86
     afterlife
    0.82
    Activations Density 0.227%

    No Known Activations