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

    Neuronpedia

    Jacobian LensNEW
    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. 43107
    Prev
    Next
    INDEX
    Explanations

    references to Local Area Networks (LAN)

    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
    326
    +0.12
    0.5%
    116
    +0.11
    0.5%
    569
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.12
    0.00
    48
    +0.11
    0.01
    116
    +0.10
    0.01
    Negative Logits
    terday
    -0.89
    lda
    -0.80
    ãĤ¡
    -0.77
    ony
    -0.77
    arians
    -0.71
    ogen
    -0.68
     Weinstein
    -0.66
    onies
    -0.65
    ogene
    -0.64
    actor
    -0.63
    POSITIVE LOGITS
    TERN
    1.20
    DS
    0.98
    TA
    0.90
    LAN
    0.85
    TABLE
    0.83
    LEY
    0.82
    HAM
    0.82
    TY
    0.81
    nect
    0.79
    STON
    0.79
    Activations Density 0.003%

    No Known Activations