© 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. 13725
    Prev
    Next
    INDEX
    Explanations

    adverbs that intensify or comment on the naturalness of a situation

    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
    455
    +0.11
    0.5%
    447
    +0.11
    0.5%
    739
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    151
    +0.11
    0.03
    318
    +0.11
    0.02
    243
    +0.10
    0.02
    Negative Logits
     lineback
    -0.71
    anmar
    -0.70
     outpost
    -0.65
    matically
    -0.64
    utenberg
    -0.64
    NING
    -0.63
    berman
    -0.63
    eday
    -0.63
     Gur
    -0.61
     Riders
    -0.61
    POSITIVE LOGITS
     provoked
    0.82
     irritated
    0.76
    ãĤ©
    0.74
     disple
    0.73
     prompted
    0.72
     angered
    0.71
     attracted
    0.71
     provoking
    0.70
     occurring
    0.70
    kward
    0.70
    Activations Density 0.044%

    No Known Activations