Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 768
    5. 8-RES_FS768-JB
    6. 61
    Prev
    Next
    INDEX
    Explanations

    names of sports teams and players

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    659
    +0.12
    0.5%
    746
    +0.11
    0.5%
    67
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.12
    0.18
    67
    +0.11
    0.16
    232
    +0.11
    0.12
    Negative Logits
    ĪĴ
    -0.72
     GOODMAN
    -0.65
     Tinder
    -0.64
     Yamaha
    -0.64
    lihood
    -0.64
    oku
    -0.63
    uyomi
    -0.62
    ãĥĦ
    -0.61
     Gors
    -0.58
     Sioux
    -0.58
    POSITIVE LOGITS
    ursive
    1.16
    urrency
    1.16
    ursor
    1.12
    rossover
    1.12
    rouch
    1.10
    overed
    1.08
    ologne
    1.07
    oding
    1.06
    URRENT
    1.06
    rescent
    1.04
    Activations Density 4.291%

    No Known Activations