© 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. 9784
    Prev
    Next
    INDEX
    Explanations

    important dates in a sequence or series

    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
    402
    +0.12
    0.5%
    622
    +0.11
    0.5%
    99
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    402
    +0.12
    0.03
    328
    +0.11
    0.02
    622
    +0.11
    0.02
    Negative Logits
    Sav
    -0.76
    rogens
    -0.72
    yers
    -0.71
    rod
    -0.71
    rs
    -0.70
    yond
    -0.70
    bara
    -0.69
    Ancient
    -0.68
    Fish
    -0.67
    lov
    -0.66
    POSITIVE LOGITS
     impressions
    1.06
     stages
    1.03
     responders
    0.96
     indications
    0.95
     impression
    0.93
    izers
    0.93
     foray
    0.92
     phases
    0.92
     reaction
    0.91
     batch
    0.89
    Activations Density 0.071%

    No Known Activations