© 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. 4482
    Prev
    Next
    INDEX
    Explanations

    phrases related to heroic sacrifices or protective actions towards others

    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
    570
    +0.12
    0.6%
    367
    +0.11
    0.5%
    34
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.12
    0.05
    34
    +0.11
    0.04
    570
    +0.11
    0.03
    Negative Logits
    sonian
    -0.67
    querque
    -0.67
    `
    -0.66
    fleet
    -0.63
    gow
    -0.60
    ngth
    -0.59
     Demand
    -0.59
    conn
    -0.59
    Jr
    -0.58
    Occup
    -0.58
    POSITIVE LOGITS
     reasons
    0.96
     example
    0.96
     awhile
    0.95
    agers
    0.95
    gotten
    0.93
     starters
    0.93
     sake
    0.92
     instance
    0.92
    aging
    0.90
     eternity
    0.89
    Activations Density 0.162%

    No Known Activations