© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 41792
    Prev
    Next
    INDEX
    Explanations

    phrases that express intensity or emphasis in statements

    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
    55
    +0.10
    0.4%
    739
    +0.10
    0.4%
    329
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.10
    0.04
    170
    +0.10
    0.03
    102
    +0.10
    0.04
    Negative Logits
     exits
    -0.76
     retrie
    -0.73
     floats
    -0.72
    stay
    -0.71
    atile
    -0.69
     roam
    -0.67
     tables
    -0.67
    stood
    -0.67
     interfaces
    -0.66
     ank
    -0.65
    POSITIVE LOGITS
     understatement
    0.94
     unacceptable
    0.89
     obviously
    0.87
     probably
    0.86
     ludicrous
    0.86
     certainly
    0.85
     folly
    0.85
     ridiculous
    0.84
     testament
    0.83
     laughable
    0.83
    Activations Density 0.088%

    No Known Activations