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. 26400
    Prev
    Next
    INDEX
    Explanations

    ordinal indicators or rankings in numerical form

    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
    373
    +0.13
    0.6%
    116
    +0.12
    0.6%
    701
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    116
    +0.13
    0.03
    701
    +0.12
    0.03
    366
    +0.10
    0.03
    Negative Logits
    worms
    -0.77
    politics
    -0.67
    rooms
    -0.67
    »Ĵ
    -0.66
    ugs
    -0.65
    helps
    -0.65
    maps
    -0.63
    zilla
    -0.63
    ãĤĭ
    -0.62
    bda
    -0.61
    POSITIVE LOGITS
     highest
    1.20
     largest
    1.20
     tallest
    1.14
     busiest
    1.12
     oldest
    1.11
     fastest
    1.09
     richest
    1.06
     cheapest
    1.06
    highest
    1.01
     smartest
    1.01
    Activations Density 0.042%

    No Known Activations