© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 6739
    Prev
    Next
    INDEX
    Explanations

    references to news articles or journalistic content from 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_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
    373
    +0.12
    0.6%
    524
    +0.10
    0.5%
    288
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    524
    +0.12
    0.02
    586
    +0.10
    0.02
    121
    +0.09
    0.02
    Negative Logits
     sediment
    -0.76
     particip
    -0.68
     timely
    -0.67
     reinvest
    -0.65
     misunderstood
    -0.65
     infiltr
    -0.63
    heartedly
    -0.63
     chilling
    -0.61
     premier
    -0.57
    ded
    -0.57
    POSITIVE LOGITS
     Images
    1.62
    sburg
    1.58
    Images
    1.17
    sburgh
    1.08
    images
    0.98
    Getty
    0.81
    photo
    0.78
    Parser
    0.77
    Photos
    0.77
    quarters
    0.77
    Activations Density 0.010%

    No Known Activations