© 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. 6814
    Prev
    Next
    INDEX
    Explanations

    phrases related to health and medical advice

    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
    608
    +0.14
    0.7%
    298
    +0.11
    0.5%
    506
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.14
    0.04
    298
    +0.11
    0.03
    666
    +0.10
    0.03
    Negative Logits
    minus
    -1.07
     ILCS
    -0.90
    arta
    -0.85
    geist
    -0.84
    idelines
    -0.82
    cles
    -0.82
    ovie
    -0.80
    notations
    -0.78
    iband
    -0.76
    arse
    -0.75
    POSITIVE LOGITS
     encountering
    1.15
     obtaining
    1.09
     acquiring
    1.05
     occurrence
    1.04
     extinction
    1.03
     survival
    1.01
     success
    0.99
     succeeding
    0.97
     miscarriage
    0.94
     becoming
    0.94
    Activations Density 0.089%

    No Known Activations