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

    Neuronpedia

    Jacobian LensNEW
    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 24576
    9. 8-RES_FS24576-JB
    11. 5646
    Prev
    Next
    INDEX
    Explanations

    news articles related to crime and accidents

    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
    640
    +0.26
    1.2%
    373
    +0.13
    0.6%
    296
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.26
    0.12
    94
    +0.13
    0.10
    591
    +0.09
    0.06
    Negative Logits
     favour
    -0.88
     scept
    -0.76
     ageing
    -0.75
     endeavour
    -0.72
     endeav
    -0.70
    utical
    -0.70
     opting
    -0.69
     organis
    -0.68
     preferring
    -0.67
     adopting
    -0.67
    POSITIVE LOGITS
    Investigators
    1.21
    Authorities
    1.20
    Police
    1.17
    Officers
    1.05
    Officials
    0.95
     Investigators
    0.94
    PHOTOS
    0.92
    Witness
    0.88
    Bus
    0.88
    Lt
    0.88
    Activations Density 0.327%

    No Known Activations