© 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. 9016
    Prev
    Next
    INDEX
    Explanations

    technology-related descriptions and reviews

    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
    481
    +0.12
    0.6%
    373
    +0.09
    0.4%
    310
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    119
    +0.12
    0.04
    271
    +0.09
    0.03
    737
    +0.09
    0.04
    Negative Logits
    rongh
    -0.79
    obook
    -0.75
    execute
    -0.73
    umbn
    -0.70
    sylvania
    -0.70
    foreseen
    -0.68
    igion
    -0.68
    ancock
    -0.68
    owship
    -0.67
    ettings
    -0.67
    POSITIVE LOGITS
     huh
    1.35
     eh
    1.24
     tho
    1.02
    !
    1.00
    !!
    0.98
    ?!
    0.94
    !?
    0.89
    !!!!
    0.89
    !)
    0.89
     ðŁĺ
    0.84
    Activations Density 0.431%

    No Known Activations