© 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. 7346
    Prev
    Next
    INDEX
    Explanations

    references to historical, political, and social events that have occurred after a certain period of time

    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
    321
    +0.11
    0.5%
    314
    +0.10
    0.5%
    540
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    321
    +0.11
    0.04
    484
    +0.10
    0.03
    540
    +0.09
    0.03
    Negative Logits
     Qiao
    -0.84
    ï¸
    -0.74
     Bett
    -0.73
     bands
    -0.70
    Ø©
    -0.69
     Orchestra
    -0.68
    aughs
    -0.67
    owship
    -0.65
     Beans
    -0.64
     Beir
    -0.63
    POSITIVE LOGITS
    mortem
    1.48
    doctoral
    1.36
    gres
    1.35
    modern
    1.27
    hum
    1.27
    graduate
    1.23
    natal
    1.21
    cards
    1.17
    election
    1.13
    greSQL
    1.13
    Activations Density 0.033%

    No Known Activations