© 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. 4802
    Prev
    Next
    INDEX
    Explanations

    words related to repeating or doing something again

    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
    748
    +0.12
    0.5%
    710
    +0.10
    0.5%
    302
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    445
    +0.12
    0.02
    748
    +0.10
    0.02
    180
    +0.10
    0.02
    Negative Logits
    uana
    -0.86
    ffe
    -0.69
    essen
    -0.65
    icial
    -0.65
    sie
    -0.65
    istical
    -0.65
    uay
    -0.65
    isu
    -0.64
    omez
    -0.64
    orea
    -0.64
    POSITIVE LOGITS
     playthrough
    0.91
    lly
    0.90
     repeats
    0.80
     offenders
    0.79
    nces
    0.78
    repeat
    0.76
    ability
    0.76
     repetition
    0.74
     repeat
    0.74
     occurrences
    0.74
    Activations Density 0.038%

    No Known Activations