© 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. 10551
    Prev
    Next
    INDEX
    Explanations

    phrases related to narratives or storytelling

    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
    37
    +0.10
    0.5%
    450
    +0.08
    0.4%
    766
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.10
    0.04
    604
    +0.08
    0.02
    105
    +0.08
    0.02
    Negative Logits
     yesterday
    -0.74
     tomorrow
    -0.74
     lately
    -0.68
     tonight
    -0.67
     Thurs
    -0.64
    formance
    -0.62
     today
    -0.61
    always
    -0.60
     currently
    -0.59
     1914
    -0.59
    POSITIVE LOGITS
     regretted
    0.72
     retract
    0.71
    gur
    0.70
     acknow
    0.68
     realise
    0.63
     apologise
    0.62
     forg
    0.61
     regret
    0.61
    OGR
    0.61
    ilial
    0.61
    Activations Density 0.237%

    No Known Activations