© 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. 367
    Prev
    Next
    INDEX
    Explanations

    references to famous works of literature or historical figures, specifically identifying the play "Macbeth."

    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
    424
    +0.12
    0.5%
    525
    +0.11
    0.5%
    89
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.12
    -0.02
    701
    +0.11
    0.04
    299
    +0.09
    0.04
    Negative Logits
    warts
    -0.72
     Marketable
    -0.68
    idan
    -0.64
     stiffness
    -0.63
    ttle
    -0.63
    gged
    -0.60
    ught
    -0.59
     detachment
    -0.58
     PATH
    -0.58
     torches
    -0.58
    POSITIVE LOGITS
    arella
    1.00
    sels
    0.89
    amia
    0.85
    arlane
    0.83
    anwhile
    0.81
    TAG
    0.79
    thur
    0.78
    arie
    0.77
     CLSID
    0.76
    atters
    0.73
    Activations Density 0.347%

    No Known Activations