© 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. 2373
    Prev
    Next
    INDEX
    Explanations

    prefixes, suffixes, and other word components that indicate an action or state

    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
    480
    +0.65
    4.1%
    87
    +0.21
    1.3%
    266
    +0.19
    1.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    266
    +0.65
    0.06
    480
    +0.21
    0.02
    232
    +0.19
    0.06
    Negative Logits
    senal
    -0.46
     initiation
    -0.45
     purse
    -0.44
    HOME
    -0.42
     intrusion
    -0.41
     Sunshine
    -0.41
     examiner
    -0.41
     treasury
    -0.41
     Paradise
    -0.40
     Settlement
    -0.40
    POSITIVE LOGITS
    antly
    0.81
    alled
    0.79
    ited
    0.78
    elled
    0.76
    ined
    0.76
    arily
    0.76
    inated
    0.76
    oused
    0.76
    ased
    0.76
    ented
    0.75
    Activations Density 0.413%

    No Known Activations