© 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. 1312
    Prev
    Next
    INDEX
    Explanations

    words related to physical sensations like denting, flinching, squinting, and shivering

    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
    78
    +0.11
    0.5%
    273
    +0.10
    0.4%
    156
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.11
    0.04
    374
    +0.10
    0.02
    526
    +0.10
    0.01
    Negative Logits
    ãĥ´
    -0.76
    OTOS
    -0.72
     Taxes
    -0.72
    ĻĤ
    -0.71
     Schools
    -0.69
    RAW
    -0.66
    Sources
    -0.66
    SE
    -0.63
    icts
    -0.63
    IRE
    -0.62
    POSITIVE LOGITS
    ingly
    1.50
    iness
    1.29
    ing
    1.26
    iest
    1.19
    y
    1.13
    ishly
    1.05
    buck
    1.05
    ings
    0.99
    ier
    0.95
    orph
    0.94
    Activations Density 0.165%

    No Known Activations