© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 3925
    Prev
    Next
    INDEX
    Explanations

    action verbs and descriptive adjectives

    oai_token-act-pair · gpt-3.5-turbo
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    380
    +0.12
    0.5%
    609
    +0.10
    0.4%
    485
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.12
    0.05
    123
    +0.10
    0.04
    526
    +0.10
    -0.00
    Negative Logits
     pitched
    -0.63
     devised
    -0.61
    arthed
    -0.61
     ALSO
    -0.61
     Nir
    -0.61
     compared
    -0.60
     authored
    -0.60
     wielded
    -0.59
     lodged
    -0.59
    oulos
    -0.59
    POSITIVE LOGITS
    pless
    1.05
    effic
    0.77
    oreal
    0.75
    odes
    0.73
    ordinate
    0.73
    roads
    0.72
    anz
    0.70
    arus
    0.69
    animate
    0.67
     parity
    0.67
    Activations Density 0.381%

    No Known Activations