© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 49152
    9. 8-RES_FS49152-JB
    11. 45772
    Prev
    Next
    INDEX
    Explanations

    instances of numerical values or quantities referenced in the text

    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
    408
    +0.12
    0.5%
    113
    +0.10
    0.5%
    377
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    43
    +0.12
    0.05
    546
    +0.10
    0.06
    765
    +0.09
    0.06
    Negative Logits
    thood
    -0.74
    obin
    -0.73
    ance
    -0.71
    ãĥīãĥ©
    -0.71
    reprene
    -0.69
    iral
    -0.69
    anish
    -0.69
    viation
    -0.69
    laughter
    -0.68
    Instant
    -0.67
    POSITIVE LOGITS
     where
    1.37
     whence
    1.20
     overlooking
    1.16
     adjoining
    1.10
     located
    1.09
     adjacent
    1.08
     situated
    1.05
     entrances
    1.05
    where
    1.04
     surrounded
    1.00
    Activations Density 0.205%

    No Known Activations