© 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 EvalsExportsinterp-engineNEWAPI 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. 5436
    Prev
    Next
    INDEX
    Explanations

    text related to crafting and sharing handmade items, specifically related to crocheting and creating patterns

    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
    255
    +0.11
    0.5%
    427
    +0.10
    0.5%
    267
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    43
    +0.11
    0.07
    297
    +0.10
    0.07
    546
    +0.10
    0.07
    Negative Logits
    ument
    -0.76
    uscript
    -0.70
    Å«
    -0.66
    Newsletter
    -0.64
    —
    -0.64
    pires
    -0.63
    olars
    -0.63
    ŃĶ
    -0.62
    itational
    -0.62
    Enlarge
    -0.61
    POSITIVE LOGITS
     namely
    1.30
     ie
    1.21
     haha
    1.17
     BUT
    1.16
     although
    1.11
     etc
    1.10
     whereas
    1.08
     Whilst
    1.05
     whilst
    1.04
     hence
    1.04
    Activations Density 0.412%

    No Known Activations