© 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. 16043
    Prev
    Next
    INDEX
    Explanations

    phrases indicating importance or significance

    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
    479
    +0.10
    0.5%
    576
    +0.10
    0.4%
    171
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    504
    +0.10
    0.02
    236
    +0.10
    0.02
    576
    +0.09
    0.01
    Negative Logits
     tolerated
    -0.62
    0000000
    -0.61
    cat
    -0.59
     independently
    -0.58
    rn
    -0.58
    nature
    -0.57
    Jewish
    -0.57
    Jenn
    -0.57
    978
    -0.57
     rav
    -0.56
    POSITIVE LOGITS
     things
    0.75
     tragedies
    0.67
     specialization
    0.66
    ij士
    0.64
     trouble
    0.64
     Leh
    0.63
     Kira
    0.63
    upon
    0.62
     Begin
    0.61
     teamwork
    0.61
    Activations Density 0.073%

    No Known Activations