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

    Neuronpedia

    Jacobian LensNEW
    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. 5976
    Prev
    Next
    INDEX
    Explanations

    first-person possessive pronouns indicating surprise or disappointment

    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
    697
    +0.10
    0.5%
    436
    +0.10
    0.4%
    663
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    436
    +0.10
    0.03
    95
    +0.10
    0.03
    298
    +0.10
    0.03
    Negative Logits
     Printed
    -0.70
    got
    -0.65
     @@
    -0.64
     Killing
    -0.64
    soDeliveryDate
    -0.61
    iatrics
    -0.57
    rolled
    -0.55
     Takes
    -0.55
     UNCLASSIFIED
    -0.55
     Directions
    -0.55
    POSITIVE LOGITS
     extent
    1.73
    venge
    1.44
     detriment
    1.32
     degree
    1.11
    rouse
    1.08
     tune
    1.03
     liking
    0.95
     discredit
    0.91
     extend
    0.86
     fullest
    0.85
    Activations Density 0.221%

    No Known Activations