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 49152
    5. 8-RES_FS49152-JB
    6. 43387
    Prev
    Next
    INDEX
    Explanations

    mentions of the day "Monday."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    430
    +0.17
    0.8%
    649
    +0.12
    0.5%
    269
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    430
    +0.17
    0.05
    384
    +0.12
    0.04
    188
    +0.10
    0.03
    Negative Logits
    ript
    -1.03
    onductor
    -0.88
    iframe
    -0.87
    ordes
    -0.78
    ribe
    -0.76
    åĬ
    -0.72
    keye
    -0.70
     Norn
    -0.70
    egal
    -0.69
    ukes
    -0.69
    POSITIVE LOGITS
     morning
    1.57
     afternoon
    1.55
     night
    1.51
     evening
    1.47
     mornings
    1.37
     Night
    1.33
    morning
    1.19
     nights
    1.18
     evenings
    1.07
    night
    0.99
    Activations Density 0.023%

    No Known Activations