© 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. 19505
    Prev
    Next
    INDEX
    Explanations

    information related to scientific research and technological advancements

    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
    87
    +0.24
    1.2%
    756
    +0.16
    0.8%
    679
    +0.15
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.24
    0.16
    679
    +0.16
    0.15
    440
    +0.15
    0.13
    Negative Logits
    ascus
    -0.75
     abduct
    -0.67
    è¦ļéĨĴ
    -0.64
    ital
    -0.63
    natureconservancy
    -0.61
     fry
    -0.60
     extinct
    -0.60
     hatch
    -0.59
    roph
    -0.59
     elig
    -0.59
    POSITIVE LOGITS
     However
    1.19
     Though
    1.18
     Although
    1.14
     Instead
    1.13
     Meanwhile
    1.13
     While
    1.12
     But
    1.12
     Those
    1.09
     Ironically
    1.06
     Initially
    1.06
    Activations Density 0.683%

    No Known Activations