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. 26361
    Prev
    Next
    INDEX
    Explanations

    instances of the word "ay."

    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
    26
    +0.10
    0.5%
    366
    +0.10
    0.5%
    709
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    26
    +0.10
    0.02
    708
    +0.10
    0.01
    284
    +0.09
    0.01
    Negative Logits
    ãĤ´ãĥ³
    -0.84
    wcs
    -0.74
    ipeg
    -0.73
     tremend
    -0.73
     Archdemon
    -0.70
    æł
    -0.68
    PsyNetMessage
    -0.68
    ugal
    -0.68
    ivity
    -0.67
    ingen
    -0.67
    POSITIVE LOGITS
    nor
    0.98
    cott
    0.96
    den
    0.86
    nard
    0.86
    alde
    0.83
    ments
    0.82
    bay
    0.81
    tes
    0.80
    walking
    0.80
    ride
    0.80
    Activations Density 0.017%

    No Known Activations