© 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 12288
    9. 8-RES_FS12288-JB
    11. 1279
    Prev
    Next
    INDEX
    Explanations

    information related to legal proceedings and government regulations

    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
    640
    +0.21
    1.0%
    373
    +0.13
    0.6%
    293
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.21
    0.15
    94
    +0.13
    0.11
    591
    +0.10
    0.09
    Negative Logits
    heit
    -0.79
     favorably
    -0.78
     favoring
    -0.69
     Palo
    -0.69
    kees
    -0.68
    boro
    -0.68
     honors
    -0.67
     utilizing
    -0.65
     favors
    -0.64
     favored
    -0.62
    POSITIVE LOGITS
    Labour
    1.14
    Scroll
    1.14
    Scotland
    0.88
    Shape
    0.85
     Trafford
    0.84
     Labour
    0.84
    Britain
    0.83
    Liverpool
    0.82
    However
    0.81
    poll
    0.80
    Activations Density 0.445%

    No Known Activations