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 3072
    5. 8-RES_FS3072-JB
    6. 1440
    Prev
    Next
    INDEX
    Explanations

    units of measurement such as distance and speed

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    545
    +0.10
    0.5%
    570
    +0.10
    0.5%
    344
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    344
    +0.10
    0.09
    238
    +0.10
    0.08
    545
    +0.10
    0.07
    Negative Logits
    eer
    -0.91
    gotten
    -0.80
    Kings
    -0.72
    effects
    -0.71
    eering
    -0.69
     Mothers
    -0.69
    mentioned
    -0.69
    udeb
    -0.68
    bred
    -0.66
    winner
    -0.66
    POSITIVE LOGITS
     apiece
    1.17
     increments
    1.04
     diameter
    0.96
    ²
    0.94
     per
    0.93
     radius
    0.91
     squared
    0.91
     (~
    0.90
     tall
    0.85
    renheit
    0.82
    Activations Density 0.586%

    No Known Activations