© 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. 14138
    Prev
    Next
    INDEX
    Explanations

    terms related to American football positions and actions

    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
    749
    +0.14
    0.6%
    339
    +0.12
    0.5%
    63
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    749
    +0.14
    0.02
    323
    +0.12
    0.02
    422
    +0.11
    0.01
    Negative Logits
     Ink
    -0.76
    auga
    -0.74
     Badge
    -0.69
    antha
    -0.69
    orporated
    -0.68
     plur
    -0.68
     Flavoring
    -0.68
    hess
    -0.68
     Brist
    -0.66
     Sadd
    -0.66
    POSITIVE LOGITS
     rusher
    0.90
    wd
    0.81
    handler
    0.80
     Passing
    0.78
    aways
    0.75
    phrase
    0.75
     blocker
    0.73
    ivity
    0.72
    ggie
    0.71
    blocking
    0.70
    Activations Density 0.024%

    No Known Activations