© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 8109
    Prev
    Next
    INDEX
    Explanations

    mentions of books, authors, and historical events or figures in the political context

    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
    393
    +0.12
    0.6%
    728
    +0.12
    0.5%
    703
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    376
    +0.12
    0.03
    728
    +0.12
    0.03
    393
    +0.10
    0.04
    Negative Logits
    avorite
    -0.81
    sylv
    -0.72
    medium
    -0.71
    \-
    -0.71
    depending
    -0.71
    Deal
    -0.70
    mitter
    -0.69
    nant
    -0.69
    oresc
    -0.68
    umerous
    -0.68
    POSITIVE LOGITS
     Other
    0.98
     Its
    0.90
     Others
    0.89
     Beyond
    0.88
     Friends
    0.87
     Politics
    0.87
     Dying
    0.87
     Mysterious
    0.84
     Transformation
    0.83
     Problem
    0.83
    Activations Density 0.183%

    No Known Activations