© 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. 549
    Prev
    Next
    INDEX
    Explanations

    sentences indicating certainty or lack of doubt

    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
    756
    +0.10
    0.5%
    455
    +0.10
    0.4%
    38
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    38
    +0.10
    0.02
    718
    +0.10
    0.02
    455
    +0.10
    0.02
    Negative Logits
     Chains
    -0.72
    ibrary
    -0.71
    ð
    -0.68
    giene
    -0.67
    ilaterally
    -0.66
    ebin
    -0.62
    eways
    -0.62
    lite
    -0.61
    agus
    -0.60
    eller
    -0.60
    POSITIVE LOGITS
     whatsoever
    0.99
    worthiness
    0.76
    ãĤª
    0.71
    cott
    0.68
     msec
    0.64
     ABOUT
    0.62
    peak
    0.62
    worthy
    0.60
     about
    0.60
     mark
    0.60
    Activations Density 0.065%

    No Known Activations