© 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. 19973
    Prev
    Next
    INDEX
    Explanations

    words related to physical restraint and fastening

    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
    43
    +0.10
    0.4%
    546
    +0.10
    0.4%
    1
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.10
    0.03
    582
    +0.10
    0.02
    526
    +0.10
    0.01
    Negative Logits
    sburg
    -0.64
    perty
    -0.62
    inces
    -0.59
    stadt
    -0.59
     WATCHED
    -0.58
    cation
    -0.58
     Carbuncle
    -0.57
    conom
    -0.57
     VIDEOS
    -0.57
    merce
    -0.57
    POSITIVE LOGITS
     onto
    1.06
     together
    1.02
     securely
    1.01
     knots
    0.95
     tightly
    0.92
    together
    0.91
     ropes
    0.87
     knot
    0.83
     up
    0.82
     shut
    0.80
    Activations Density 0.124%

    No Known Activations