© 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. 15490
    Prev
    Next
    INDEX
    Explanations

    phrases related to announcing and discussing reports or investigation results

    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
    616
    +0.11
    0.5%
    206
    +0.10
    0.4%
    291
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.11
    0.03
    608
    +0.10
    0.03
    616
    +0.09
    0.03
    Negative Logits
    ppa
    -0.71
    caps
    -0.70
    pload
    -0.67
    âĺ
    -0.67
    ospace
    -0.67
    quit
    -0.67
    cko
    -0.66
    head
    -0.66
    forward
    -0.65
    afety
    -0.65
    POSITIVE LOGITS
     experimentation
    0.84
     deliberations
    0.80
     negotiations
    0.79
     experiments
    0.78
     investigation
    0.77
     exhaustive
    0.77
     testing
    0.75
     surgeries
    0.74
     painstaking
    0.74
     investigations
    0.73
    Activations Density 0.123%

    No Known Activations