© 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. 10280
    Prev
    Next
    INDEX
    Explanations

    words related to sports events or competitions

    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
    447
    +0.24
    1.1%
    680
    +0.11
    0.5%
    361
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    651
    +0.24
    0.07
    540
    +0.11
    0.07
    288
    +0.11
    0.07
    Negative Logits
     abound
    -0.70
    yrics
    -0.69
    wolves
    -0.64
    entimes
    -0.58
    ixtures
    -0.58
    nature
    -0.57
    itability
    -0.56
    wealth
    -0.55
     pathological
    -0.54
     nerves
    -0.54
    POSITIVE LOGITS
    imester
    1.00
     installment
    0.92
     iteration
    0.78
    alion
    0.78
     installments
    0.75
    rant
    0.74
    anza
    0.74
     incarnation
    0.70
    icipated
    0.69
     outing
    0.69
    Activations Density 0.504%

    No Known Activations