© 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 49152
    9. 8-RES_FS49152-JB
    11. 26563
    Prev
    Next
    INDEX
    Explanations

    variations of the verb "be."

    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
    393
    +0.12
    0.6%
    55
    +0.12
    0.6%
    102
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.12
    0.06
    102
    +0.12
    0.06
    7
    +0.12
    0.04
    Negative Logits
    rix
    -0.68
    PHOTOS
    -0.68
    odore
    -0.68
    pedia
    -0.65
     Notting
    -0.65
    riott
    -0.62
     Muse
    -0.62
    rones
    -0.61
     Tigers
    -0.61
     Morton
    -0.60
    POSITIVE LOGITS
    fall
    1.15
    fits
    1.12
    fit
    1.11
    fitting
    0.97
     able
    0.97
    falls
    0.95
     construed
    0.95
     considered
    0.91
     termed
    0.90
     incorporated
    0.88
    Activations Density 0.094%

    No Known Activations