© 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. 10280
    Prev
    Next
    INDEX
    Explanations

    phrases containing the word "in"

    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
    640
    +0.15
    0.7%
    121
    +0.12
    0.6%
    314
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    121
    +0.15
    0.04
    313
    +0.12
    0.04
    235
    +0.11
    0.04
    Negative Logits
    76561
    -0.73
     %%
    -0.66
     egreg
    -0.63
    killed
    -0.63
     combatants
    -0.63
     manip
    -0.63
     needles
    -0.61
     snipers
    -0.61
     needs
    -0.61
     unnecess
    -0.59
    POSITIVE LOGITS
     unison
    1.13
     lieu
    1.01
    clus
    1.00
     response
    0.94
     conjunction
    0.89
     advance
    0.89
     accordance
    0.87
    humane
    0.87
     regards
    0.87
     order
    0.86
    Activations Density 0.073%

    No Known Activations