© 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 EvalsExportsinterp-engineNEWAPI 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. 5343
    Prev
    Next
    INDEX
    Explanations

    phrases related to providing explanations and instructions

    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
    228
    +0.10
    0.4%
    262
    +0.09
    0.4%
    559
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    559
    +0.10
    0.01
    214
    +0.09
    0.01
    172
    +0.09
    0.01
    Negative Logits
     rave
    -0.76
    wagon
    -0.66
    tsy
    -0.64
     Growth
    -0.56
     seriousness
    -0.56
     trillion
    -0.56
    ATIONAL
    -0.55
    ships
    -0.55
    IVE
    -0.54
     wonder
    -0.52
    POSITIVE LOGITS
    ynchron
    0.75
    é¾įåĸļ士
    0.72
    ociated
    0.67
    orah
    0.64
    hest
    0.63
     efficiently
    0.63
    velt
    0.63
     instructed
    0.63
    perse
    0.63
    ç¥ŀ
    0.63
    Activations Density 0.034%

    No Known Activations