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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 1607
    Prev
    Next
    INDEX
    Explanations

    legal terms and trial-related phrases

    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
    447
    +0.12
    0.5%
    288
    +0.10
    0.5%
    488
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    500
    +0.12
    0.02
    268
    +0.10
    0.02
    488
    +0.10
    0.02
    Negative Logits
    otle
    -0.76
    oji
    -0.75
    kell
    -0.70
    erd
    -0.68
    mson
    -0.67
    orns
    -0.66
    ula
    -0.65
    amera
    -0.64
    lav
    -0.63
    licks
    -0.62
    POSITIVE LOGITS
     proceedings
    0.89
     Proceedings
    0.89
     hearings
    0.89
    Court
    0.86
    eering
    0.82
     Trials
    0.81
     transcripts
    0.80
     testimony
    0.80
    eer
    0.80
     judge
    0.80
    Activations Density 0.024%

    No Known Activations