© 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. 41603
    Prev
    Next
    INDEX
    Explanations

    phrases indicating observations or notable information

    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
    617
    +0.10
    0.4%
    690
    +0.09
    0.4%
    457
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    690
    +0.10
    0.03
    13
    +0.09
    0.02
    125
    +0.09
    0.02
    Negative Logits
     mole
    -0.62
    holes
    -0.59
    abouts
    -0.59
     Crescent
    -0.57
    ãĥı
    -0.56
     Trib
    -0.56
     sightings
    -0.56
    ãĥŃ
    -0.55
    "},"
    -0.54
     Dollar
    -0.54
    POSITIVE LOGITS
     clarify
    0.85
     borne
    0.81
    erest
    0.79
     emphas
    0.77
    antz
    0.77
     abundantly
    0.74
     emphasized
    0.72
    bnb
    0.71
     clar
    0.71
    pec
    0.71
    Activations Density 0.065%

    No Known Activations