© 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
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 12288
    9. 8-RES_FS12288-JB
    11. 1752
    Prev
    Next
    INDEX
    Explanations

    locations and events related to conflict and crisis

    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
    180
    +0.12
    0.5%
    362
    +0.10
    0.5%
    181
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    180
    +0.12
    0.03
    611
    +0.10
    0.03
    669
    +0.09
    0.03
    Negative Logits
    appropriate
    -0.73
     Cosponsors
    -0.70
    Http
    -0.70
    Choice
    -0.67
    Subject
    -0.67
    Versions
    -0.66
     istg
    -0.66
    Own
    -0.65
    Password
    -0.64
    20439
    -0.64
    POSITIVE LOGITS
     Baghdad
    0.97
     Bethlehem
    0.93
     Aleppo
    0.87
     Lima
    0.86
     Mosul
    0.86
     Brisbane
    0.86
     Johannes
    0.85
     Perth
    0.83
     Louisville
    0.83
     Mexico
    0.83
    Activations Density 0.106%

    No Known Activations