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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 19051
    Prev
    Next
    INDEX
    Explanations

    the phrase "as soon as."

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    348
    +0.10
    0.5%
    98
    +0.10
    0.4%
    271
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    98
    +0.10
    0.01
    348
    +0.10
    0.01
    172
    +0.09
    0.01
    Negative Logits
    ricks
    -0.79
    roo
    -0.71
    uph
    -0.67
    âĹ¼
    -0.67
    sson
    -0.64
    bold
    -0.63
    arnaev
    -0.63
    ARS
    -0.62
    tie
    -0.62
    tw
    -0.62
    POSITIVE LOGITS
     practicable
    0.95
     possible
    0.87
     they
    0.78
     clicked
    0.76
     payday
    0.75
     dawn
    0.73
     whiff
    0.70
     someone
    0.70
     somebody
    0.69
     sunrise
    0.69
    Activations Density 0.019%

    No Known Activations