© 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. 16939
    Prev
    Next
    INDEX
    Explanations

    verbs related to actions or tasks that emphasize a sense of importance or seriousness

    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
    78
    +0.10
    0.4%
    458
    +0.09
    0.4%
    698
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    78
    +0.10
    0.04
    588
    +0.09
    0.04
    698
    +0.09
    0.03
    Negative Logits
     Entered
    -0.76
     Det
    -0.62
     suspects
    -0.62
     Gork
    -0.61
    hog
    -0.60
     tongues
    -0.59
    mares
    -0.59
    ussen
    -0.58
     boarded
    -0.56
    rules
    -0.56
    POSITIVE LOGITS
    pez
    1.02
     differently
    0.86
     administr
    0.79
     wrong
    0.79
    女
    0.78
    ãĥ¼ãĤ¯
    0.73
    igraph
    0.72
    ournal
    0.71
    hing
    0.71
    etting
    0.70
    Activations Density 0.046%

    No Known Activations