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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 8328
    Prev
    Next
    INDEX
    Explanations

    times and locations within a story

    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
    371
    +0.12
    0.5%
    362
    +0.11
    0.5%
    676
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    371
    +0.12
    0.07
    362
    +0.11
    0.06
    172
    +0.11
    0.05
    Negative Logits
     benefited
    -0.70
    aceutical
    -0.69
    gans
    -0.66
    eria
    -0.66
    uph
    -0.62
    PH
    -0.61
    tg
    -0.59
    llor
    -0.59
    spective
    -0.58
    oulos
    -0.58
    POSITIVE LOGITS
     least
    1.28
     sunset
    0.99
     halftime
    0.97
     dusk
    0.93
    onement
    0.91
    yp
    0.91
     sunrise
    0.89
     midnight
    0.89
     midday
    0.89
     noon
    0.89
    Activations Density 0.156%

    No Known Activations