© 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. 13668
    Prev
    Next
    INDEX
    Explanations

    technical terms and information related to government legislation and policies

    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
    +0.14
    0.7%
    152
    +0.10
    0.5%
    102
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    152
    +0.14
    0.03
    416
    +0.10
    0.03
    619
    +0.10
    0.03
    Negative Logits
     shrink
    -0.63
     revision
    -0.63
     Rivals
    -0.61
     Lumpur
    -0.60
     Borders
    -0.59
    ONEY
    -0.58
     Gillespie
    -0.58
     Reloaded
    -0.57
     Stream
    -0.57
     argument
    -0.57
    POSITIVE LOGITS
    purpose
    1.23
    important
    1.14
    star
    1.06
    powerful
    1.06
    season
    1.01
    consuming
    1.00
    winner
    1.00
    around
    1.00
    access
    0.98
    rounder
    0.97
    Activations Density 0.019%

    No Known Activations