© 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 24576
    9. 8-RES_FS24576-JB
    11. 1561
    Prev
    Next
    INDEX
    Explanations

    mentions of political parties, particularly Democrats

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    +0.17
    0.8%
    282
    +0.11
    0.5%
    488
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    488
    +0.17
    0.04
    282
    +0.11
    0.03
    432
    +0.11
    0.03
    Negative Logits
    iful
    -0.84
     Jehovah
    -0.71
    istic
    -0.71
    isu
    -0.70
    obar
    -0.67
    ãĤ¨ãĥ«
    -0.67
    aji
    -0.66
    ļé
    -0.66
    uras
    -0.65
    azing
    -0.65
    POSITIVE LOGITS
    pace
    1.03
    paces
    1.00
    hift
    0.93
    ervatives
    0.92
    hip
    0.88
    ervative
    0.85
     accuse
    0.84
     nominate
    0.83
    '
    0.79
    lash
    0.78
    Activations Density 0.060%

    No Known Activations