© 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. 744
    Prev
    Next
    INDEX
    Explanations

    statements about the utilization of certain tools or practices

    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
    554
    +0.12
    0.6%
    420
    +0.12
    0.5%
    686
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    554
    +0.12
    0.02
    677
    +0.12
    0.04
    686
    +0.10
    0.04
    Negative Logits
     Hai
    -0.73
    bitious
    -0.70
    alos
    -0.62
    nesty
    -0.62
     undertaking
    -0.61
     Roose
    -0.60
    ipeg
    -0.58
     achie
    -0.58
    gdala
    -0.57
    merce
    -0.57
    POSITIVE LOGITS
     sparing
    1.13
    fully
    0.98
     interchange
    0.92
     extensively
    0.84
    FUL
    0.79
     exclusively
    0.77
     captcha
    0.73
    pez
    0.73
     wisely
    0.72
    ragon
    0.70
    Activations Density 0.095%

    No Known Activations