© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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 12288
    9. 8-RES_FS12288-JB
    11. 10840
    Prev
    Next
    INDEX
    Explanations

    situations related to criminal activities

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    373
    +0.11
    0.5%
    121
    +0.10
    0.5%
    723
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    121
    +0.11
    0.06
    723
    +0.10
    0.05
    105
    +0.10
    0.04
    Negative Logits
     contrace
    -0.84
     filib
    -0.77
    gemony
    -0.71
    lip
    -0.69
    bench
    -0.67
     lect
    -0.67
     nostalg
    -0.66
     percept
    -0.65
    issance
    -0.65
     subord
    -0.64
    POSITIVE LOGITS
     Investigators
    1.06
    Investigators
    1.05
    arnaev
    1.03
    Authorities
    1.00
     detectives
    0.97
     coroner
    0.96
     Authorities
    0.96
    nsics
    0.93
     investigators
    0.90
     COUNTY
    0.89
    Activations Density 0.713%

    No Known Activations