© 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. 17435
    Prev
    Next
    INDEX
    Explanations

    questions and statements related to inquiry or investigation, specifically involving people's actions, opinions, or beliefs

    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
    374
    +0.12
    0.5%
    560
    +0.11
    0.5%
    693
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    60
    +0.12
    0.03
    374
    +0.11
    0.02
    287
    +0.10
    0.02
    Negative Logits
    ulic
    -0.86
    tails
    -0.74
    WT
    -0.74
    imum
    -0.66
    tes
    -0.66
    ymph
    -0.65
    ahn
    -0.65
    eps
    -0.65
    rawdownloadcloneembedreportprint
    -0.64
    ε
    -0.64
    POSITIVE LOGITS
     recollection
    0.84
     recol
    0.77
     hes
    0.75
     forgiveness
    0.74
     specifics
    0.72
     bothered
    0.71
     hesitation
    0.70
     congratulations
    0.70
     possible
    0.67
     objections
    0.67
    Activations Density 0.099%

    No Known Activations