© 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. 13057
    Prev
    Next
    INDEX
    Explanations

    personal pronouns and connection words indicating interaction and personal narratives

    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
    373
    +0.14
    0.7%
    483
    +0.10
    0.5%
    134
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    483
    +0.14
    0.03
    628
    +0.10
    0.03
    307
    +0.10
    0.03
    Negative Logits
    xtap
    -0.67
    hift
    -0.61
    ibal
    -0.61
    aired
    -0.61
    arious
    -0.61
    arij
    -0.60
    aband
    -0.60
    HOU
    -0.60
     Wikimedia
    -0.59
    hement
    -0.59
    POSITIVE LOGITS
     beforehand
    0.91
     how
    0.84
     goodbye
    0.84
     bluntly
    0.81
     plainly
    0.80
     about
    0.78
     why
    0.78
     orally
    0.77
     otherwise
    0.75
    DERR
    0.71
    Activations Density 0.068%

    No Known Activations