© 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. 15727
    Prev
    Next
    INDEX
    Explanations

    words related to languages and specific language terms

    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
    549
    +0.11
    0.5%
    314
    +0.09
    0.4%
    409
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.02
    409
    +0.09
    0.01
    57
    +0.09
    0.01
    Negative Logits
     externalToEVAOnly
    -0.94
    scene
    -0.71
     CTR
    -0.70
    */(
    -0.70
    endez
    -0.69
     Parables
    -0.68
    displayText
    -0.68
    osponsors
    -0.68
     HRC
    -0.68
    bub
    -0.67
    POSITIVE LOGITS
    Sax
    0.97
     Anglo
    0.93
    ixture
    0.78
    wegian
    0.78
    Indian
    0.75
     Sax
    0.73
    wealth
    0.73
    volent
    0.72
    ghan
    0.70
     Celt
    0.70
    Activations Density 0.032%

    No Known Activations