© 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. 14776
    Prev
    Next
    INDEX
    Explanations

    information related to criminal activities and seeking help or tips from the public

    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
    726
    +0.12
    0.5%
    639
    +0.12
    0.5%
    40
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    726
    +0.12
    0.03
    424
    +0.12
    0.02
    639
    +0.12
    0.02
    Negative Logits
    gemony
    -0.79
    yt
    -0.70
    Rated
    -0.69
     guiActiveUnfocused
    -0.69
     Democr
    -0.67
    chin
    -0.66
     subord
    -0.66
    Zen
    -0.64
     Balanced
    -0.62
    mostly
    -0.61
    POSITIVE LOGITS
     whereabouts
    0.95
     enqu
    0.91
     inquiries
    0.87
    iries
    0.85
     information
    0.83
     hotline
    0.83
     sightings
    0.82
     emergencies
    0.81
     contact
    0.80
     confidential
    0.79
    Activations Density 0.098%

    No Known Activations