© 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. 4915
    Prev
    Next
    INDEX
    Explanations

    information related to online comments and website interactions

    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
    373
    +0.16
    0.8%
    640
    +0.14
    0.7%
    200
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.16
    0.10
    373
    +0.14
    -0.02
    143
    +0.10
    0.05
    Negative Logits
    paces
    -0.95
    hement
    -0.73
    recy
    -0.70
    tk
    -0.69
    cius
    -0.67
    IAS
    -0.66
    cuts
    -0.65
     Cent
    -0.63
    abouts
    -0.63
    being
    -0.60
    POSITIVE LOGITS
     chimpan
    0.55
     saliva
    0.54
     triggers
    0.53
    ixture
    0.53
     grips
    0.53
     dispatcher
    0.53
     provoke
    0.52
     genetic
    0.52
     rapt
    0.52
    BuyableInstoreAndOnline
    0.52
    Activations Density 0.272%

    No Known Activations