© 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. 2698
    Prev
    Next
    INDEX
    Explanations

    ideas and phrases related to historical and philosophical discussions

    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
    25
    +0.11
    0.5%
    366
    +0.10
    0.5%
    549
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    25
    +0.11
    0.06
    609
    +0.10
    0.06
    366
    +0.10
    0.04
    Negative Logits
    oof
    -0.69
    ulner
    -0.67
    ories
    -0.65
    ooks
    -0.65
    ãĥ©ãĥ³
    -0.65
    emies
    -0.64
    thumbnails
    -0.63
    photos
    -0.62
    osponsors
    -0.62
    akings
    -0.62
    POSITIVE LOGITS
     hundred
    1.08
     thousand
    0.96
     assumes
    0.92
     Hundred
    0.92
     thing
    0.89
     wonders
    0.86
     learns
    0.83
     person
    0.82
     sided
    0.81
     cannot
    0.76
    Activations Density 0.108%

    No Known Activations