© 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. 5452
    Prev
    Next
    INDEX
    Explanations

    numeric values related to width in a specific context

    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
    99
    +0.16
    0.7%
    526
    +0.12
    0.5%
    365
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.16
    0.03
    326
    +0.12
    0.02
    288
    +0.11
    0.03
    Negative Logits
    uate
    -0.75
    ogical
    -0.74
    nces
    -0.72
    amount
    -0.72
    ucl
    -0.71
    ally
    -0.70
    ament
    -0.69
    abolic
    -0.69
    abol
    -0.69
    ateur
    -0.68
    POSITIVE LOGITS
    sheet
    0.94
    sburg
    0.82
    s
    0.81
    ï¸
    0.78
    heet
    0.78
    ington
    0.74
    strings
    0.71
    swe
    0.69
    sheets
    0.68
    ãģį
    0.68
    Activations Density 0.059%

    No Known Activations