© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 1792
    Prev
    Next
    INDEX
    Explanations

    descriptions related to childhood memories and experiences

    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
    132
    +0.11
    0.5%
    102
    +0.11
    0.5%
    235
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    132
    +0.11
    0.04
    356
    +0.11
    0.03
    545
    +0.11
    0.03
    Negative Logits
    ioxide
    -0.76
    atility
    -0.70
     Desk
    -0.68
    cific
    -0.68
    aminer
    -0.67
     vacancy
    -0.63
     unfocusedRange
    -0.63
    idine
    -0.63
     contingency
    -0.62
     POS
    -0.62
    POSITIVE LOGITS
    digy
    0.95
     idol
    0.87
     boy
    0.86
    boy
    0.85
     boys
    0.84
    kid
    0.82
     summers
    0.78
     puberty
    0.77
     Gohan
    0.77
    kids
    0.77
    Activations Density 0.121%

    No Known Activations