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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 46106
    Prev
    Next
    INDEX
    Explanations

    references to projects and their statuses, particularly regarding construction and authorization

    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
    72
    +0.12
    0.6%
    205
    +0.11
    0.5%
    584
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.12
    0.05
    690
    +0.11
    0.04
    566
    +0.11
    0.03
    Negative Logits
    erald
    -0.71
     representations
    -0.71
    washed
    -0.68
    deen
    -0.67
     substitutes
    -0.67
     Column
    -0.66
    lasses
    -0.66
     Plain
    -0.65
     Glac
    -0.65
     lashes
    -0.64
    POSITIVE LOGITS
     undertaken
    1.46
     funded
    1.31
     underway
    1.27
     financed
    1.24
     feasibility
    1.05
     spearheaded
    1.04
    funded
    1.03
    ONSORED
    0.98
    pleted
    0.97
     planned
    0.97
    Activations Density 0.316%

    No Known Activations