© 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. 4271
    Prev
    Next
    INDEX
    Explanations

    words related to personal reflections and self-improvement

    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
    481
    +0.12
    0.6%
    377
    +0.12
    0.5%
    291
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.12
    0.05
    377
    +0.12
    0.04
    55
    +0.10
    0.04
    Negative Logits
     ban
    -0.61
     lur
    -0.61
     Frazier
    -0.60
    uish
    -0.58
     Bac
    -0.58
     Baal
    -0.57
    iencies
    -0.57
     Chan
    -0.56
     bans
    -0.56
    teness
    -0.56
    POSITIVE LOGITS
     reacting
    0.89
     performing
    0.86
    abouts
    0.83
     interacting
    0.82
     enjoying
    0.82
    senal
    0.82
     behaving
    0.81
     experiencing
    0.78
    é¾įåĸļ士
    0.78
     heading
    0.77
    Activations Density 0.104%

    No Known Activations