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. 46078
    Prev
    Next
    INDEX
    Explanations

    questions seeking opinions, reflections, and insights

    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
    27
    +0.11
    0.5%
    60
    +0.10
    0.5%
    497
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    60
    +0.11
    0.04
    449
    +0.10
    0.03
    434
    +0.10
    0.03
    Negative Logits
    oys
    -0.69
    plain
    -0.67
    cession
    -0.67
    Marx
    -0.66
    pool
    -0.64
    $$$$
    -0.63
    $$
    -0.63
    fort
    -0.63
    Merc
    -0.63
    printf
    -0.63
    POSITIVE LOGITS
     recollection
    0.77
     memories
    0.77
    ?ãĢį
    0.74
     memorable
    0.73
     experien
    0.73
     accomp
    0.70
     impressions
    0.70
     misconceptions
    0.68
    Favorite
    0.68
    Desc
    0.67
    Activations Density 0.090%

    No Known Activations