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. 41314
    Prev
    Next
    INDEX
    Explanations

    references to the specific location "Crisp"

    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
    118
    +0.09
    0.4%
    611
    +0.09
    0.4%
    195
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    118
    +0.09
    0.01
    232
    +0.09
    0.01
    348
    +0.09
    0.01
    Negative Logits
    izer
    -0.70
     Antar
    -0.68
     chant
    -0.65
     buoy
    -0.63
     gaze
    -0.62
    Ku
    -0.62
     Sabha
    -0.61
     laundering
    -0.61
    meal
    -0.59
    gdala
    -0.59
    POSITIVE LOGITS
    erella
    1.17
    sburgh
    1.09
    shire
    0.90
    erey
    0.89
    sylvania
    0.88
    s
    0.86
    legate
    0.85
    orld
    0.84
    Berry
    0.79
    bone
    0.79
    Activations Density 0.003%

    No Known Activations