© 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. 10287
    Prev
    Next
    INDEX
    Explanations

    references to images or photos credited to Getty Images

    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
    373
    +0.11
    0.5%
    524
    +0.10
    0.4%
    288
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    586
    +0.11
    0.03
    524
    +0.10
    0.03
    121
    +0.09
    0.03
    Negative Logits
     sediment
    -0.73
     misunderstood
    -0.67
    heartedly
    -0.65
     reinvest
    -0.65
     infiltr
    -0.65
     timely
    -0.64
     particip
    -0.64
    stood
    -0.64
     unreal
    -0.61
     vain
    -0.60
    POSITIVE LOGITS
     Images
    1.69
    sburg
    1.64
    Images
    1.20
    sburgh
    1.06
    images
    1.00
     IMAGES
    0.84
    Photos
    0.82
    Getty
    0.82
    photo
    0.81
    Film
    0.81
    Activations Density 0.038%

    No Known Activations