© 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. 1647
    Prev
    Next
    INDEX
    Explanations

    questions and discussions related to privacy, government surveillance, and social justice

    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
    192
    +0.13
    0.6%
    373
    +0.10
    0.5%
    356
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    449
    +0.13
    0.04
    287
    +0.10
    0.03
    643
    +0.09
    0.03
    Negative Logits
    places
    -0.72
    chairs
    -0.71
    gra
    -0.65
    ties
    -0.63
    lly
    -0.63
    chat
    -0.62
    forts
    -0.61
    irds
    -0.61
    verage
    -0.59
    values
    -0.58
    POSITIVE LOGITS
     Thou
    0.78
     thou
    0.77
     anybody
    0.74
     there
    0.73
     YOU
    0.70
     you
    0.69
     ya
    0.68
     anyone
    0.67
     THEY
    0.66
    yout
    0.66
    Activations Density 0.126%

    No Known Activations