© 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. 10315
    Prev
    Next
    INDEX
    Explanations

    words related to winter and seasonal changes

    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
    617
    +0.13
    0.6%
    554
    +0.12
    0.5%
    575
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    617
    +0.13
    0.02
    575
    +0.12
    0.02
    571
    +0.11
    0.02
    Negative Logits
    Pages
    -0.75
    ocrat
    -0.74
    lesi
    -0.73
    ocratic
    -0.73
    ifice
    -0.72
    angler
    -0.72
    raltar
    -0.72
    aurus
    -0.71
    reme
    -0.70
    ronics
    -0.70
    POSITIVE LOGITS
    thur
    1.07
     Wonderland
    0.99
     coats
    0.87
    time
    0.85
    fell
    0.85
     sol
    0.85
    flake
    0.81
     mornings
    0.81
    hawks
    0.78
     rains
    0.78
    Activations Density 0.049%

    No Known Activations