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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 26452
    Prev
    Next
    INDEX
    Explanations

    occurrences of the name "Jerry."

    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
    430
    +0.15
    0.7%
    419
    +0.10
    0.4%
    26
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    430
    +0.15
    0.01
    26
    +0.10
    0.01
    419
    +0.10
    0.01
    Negative Logits
    ngth
    -0.82
    atively
    -0.72
    alid
    -0.72
    awaru
    -0.71
    hips
    -0.71
    yrim
    -0.69
     totality
    -0.68
    heed
    -0.68
    hip
    -0.66
    ebin
    -0.66
    POSITIVE LOGITS
     Springer
    1.09
     Kramer
    0.89
     Coy
    0.76
    stein
    0.75
     Fal
    0.75
    ono
    0.74
     Angelo
    0.73
     Garcia
    0.71
     Reese
    0.70
     Lee
    0.70
    Activations Density 0.004%

    No Known Activations