© 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. 1603
    Prev
    Next
    INDEX
    Explanations

    references to specific locations or entities

    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
    288
    +0.14
    0.7%
    99
    +0.13
    0.6%
    743
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.14
    0.02
    219
    +0.13
    0.02
    694
    +0.10
    0.01
    Negative Logits
    ovie
    -0.87
    hift
    -0.79
    omo
    -0.77
    ospace
    -0.74
    kson
    -0.73
    phant
    -0.71
    fm
    -0.71
    actic
    -0.69
    iatric
    -0.69
    eger
    -0.68
    POSITIVE LOGITS
     Beach
    1.41
     Tech
    1.30
     Commonwealth
    1.01
     Wool
    0.91
     Dare
    0.89
     Springs
    0.87
     Mayo
    0.84
     Roads
    0.84
     Hills
    0.84
     Oaks
    0.79
    Activations Density 0.017%

    No Known Activations