© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 8137
    Prev
    Next
    INDEX
    Explanations

    mentions or discussions of political moderation

    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
    496
    +0.13
    0.6%
    195
    +0.11
    0.5%
    116
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    45
    +0.13
    0.01
    528
    +0.11
    0.01
    496
    +0.09
    0.02
    Negative Logits
    arium
    -0.73
    chy
    -0.71
    stals
    -0.70
    atography
    -0.69
    sten
    -0.68
    raltar
    -0.68
    kefeller
    -0.68
    borne
    -0.67
    OGR
    -0.66
    Report
    -0.65
    POSITIVE LOGITS
    erate
    1.07
     sized
    0.99
     minded
    0.97
    xual
    0.93
     (<
    0.80
    lees
    0.70
     leaning
    0.69
     (~
    0.69
     medi
    0.69
    moderate
    0.68
    Activations Density 0.032%

    No Known Activations