© 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. 45740
    Prev
    Next
    INDEX
    Explanations

    sentences with punctuation marks, indicating possible conclusions or significant statements

    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
    459
    +0.18
    0.9%
    497
    +0.14
    0.7%
    496
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.18
    0.09
    679
    +0.14
    0.08
    454
    +0.11
    0.07
    Negative Logits
     skelet
    -0.77
    lihood
    -0.70
    pled
    -0.70
    onite
    -0.68
    bage
    -0.68
     unden
    -0.67
    ¥ŀ
    -0.66
     fung
    -0.65
     sack
    -0.65
     attrition
    -0.65
    POSITIVE LOGITS
     Specifically
    0.99
     Yesterday
    0.94
     Turns
    0.92
     Drawing
    0.85
     Transcript
    0.84
     Known
    0.84
     Whether
    0.84
     Previously
    0.83
     Related
    0.83
     Apparently
    0.82
    Activations Density 0.218%

    No Known Activations