© 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. 3786
    Prev
    Next
    INDEX
    Explanations

    information related to reporting crimes and incidents to police

    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
    639
    +0.11
    0.5%
    511
    +0.10
    0.4%
    5
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    639
    +0.11
    0.05
    726
    +0.10
    0.04
    380
    +0.09
    0.03
    Negative Logits
    puff
    -0.84
    gemony
    -0.79
     stacked
    -0.75
     dominates
    -0.73
     feats
    -0.71
     lavish
    -0.70
     bip
    -0.69
     enriched
    -0.69
     hips
    -0.69
     layouts
    -0.68
    POSITIVE LOGITS
     hotline
    1.46
    Emergency
    1.09
    police
    1.07
     emergency
    1.03
     RCMP
    1.01
     paramedics
    1.00
    Police
    0.99
     enqu
    0.99
     ambulance
    0.98
     emergencies
    0.98
    Activations Density 0.672%

    No Known Activations