© 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. 19981
    Prev
    Next
    INDEX
    Explanations

    technical descriptions of equipment and facilities

    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
    664
    +0.13
    0.6%
    108
    +0.10
    0.5%
    10
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    664
    +0.13
    0.04
    206
    +0.10
    0.03
    59
    +0.09
    0.03
    Negative Logits
    chy
    -0.82
    esame
    -0.72
    jab
    -0.70
    browser
    -0.69
    itsch
    -0.69
    TPP
    -0.68
    isman
    -0.66
    é»Ĵ
    -0.64
    lla
    -0.64
    bell
    -0.63
    POSITIVE LOGITS
     larger
    0.78
     occupants
    0.77
     multiple
    0.74
     incoming
    0.72
     capacities
    0.71
     everything
    0.71
     occupancy
    0.71
     unlimited
    0.70
     thousands
    0.69
     3000
    0.69
    Activations Density 0.143%

    No Known Activations