© 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 12288
    9. 8-RES_FS12288-JB
    11. 834
    Prev
    Next
    INDEX
    Explanations

    specific names and technical 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_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    447
    +0.32
    1.5%
    490
    +0.12
    0.5%
    650
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.32
    0.07
    461
    +0.12
    0.06
    288
    +0.10
    0.08
    Negative Logits
    abwe
    -0.74
    adows
    -0.69
    sembly
    -0.69
    licts
    -0.68
    ĥ
    -0.68
    OHN
    -0.67
    idth
    -0.67
    į
    -0.67
    anchester
    -0.66
    oos
    -0.66
    POSITIVE LOGITS
    's
    0.85
     Labs
    0.75
     itself
    0.72
     himself
    0.64
    '
    0.61
     herself
    0.61
    ->
    0.60
     publishes
    0.59
     Industries
    0.59
    ::
    0.59
    Activations Density 0.574%

    No Known Activations