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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 37103
    Prev
    Next
    INDEX
    Explanations

    sentences that express strong emotion or feelings

    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
    6
    +0.13
    0.6%
    640
    +0.13
    0.6%
    430
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.13
    0.05
    267
    +0.13
    0.04
    248
    +0.12
    0.03
    Negative Logits
    ensibly
    -0.79
    apers
    -0.75
    graded
    -0.73
    aper
    -0.73
    ciplinary
    -0.72
    ¥ŀ
    -0.72
    undai
    -0.71
    azo
    -0.71
    userc
    -0.71
    asers
    -0.70
    POSITIVE LOGITS
     âĢķ
    0.98
     Asked
    0.93
    [/
    0.91
    <|endoftext|>
    0.86
     Adds
    0.84
     Cue
    0.82
     ["
    0.79
     Then
    0.78
     Hearing
    0.77
     Lastly
    0.77
    Activations Density 0.055%

    No Known Activations