© 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 24576
    9. 8-RES_FS24576-JB
    11. 13023
    Prev
    Next
    INDEX
    Explanations

    detailed descriptions or narratives explaining complex scenarios or dilemmas

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    167
    +0.12
    0.5%
    578
    +0.11
    0.5%
    11
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.12
    0.06
    537
    +0.11
    0.05
    11
    +0.10
    0.03
    Negative Logits
     Sweeney
    -0.68
    Thompson
    -0.66
    storm
    -0.63
     HuffPost
    -0.60
    llo
    -0.59
    aneously
    -0.59
     Sau
    -0.58
     Wasserman
    -0.58
     Thompson
    -0.57
     Deng
    -0.57
    POSITIVE LOGITS
    hip
    1.35
    ilver
    1.24
    omething
    1.21
    chool
    1.21
    mith
    1.19
    peed
    1.14
    erver
    1.14
    kaya
    1.12
    pace
    1.11
    hips
    1.11
    Activations Density 0.367%

    No Known Activations