© 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. 18452
    Prev
    Next
    INDEX
    Explanations

    mentions of outdoor activities and adventures

    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
    269
    +0.13
    0.6%
    90
    +0.12
    0.5%
    265
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    269
    +0.13
    0.07
    90
    +0.12
    0.07
    499
    +0.10
    0.05
    Negative Logits
    oland
    -0.64
     reprinted
    -0.63
    anan
    -0.61
    Dist
    -0.61
    imilation
    -0.61
    Definition
    -0.60
    ollah
    -0.60
    abo
    -0.60
    cellence
    -0.60
    ,,
    -0.58
    POSITIVE LOGITS
     behalf
    1.58
     occasion
    1.12
    etime
    1.11
    shore
    1.09
    stage
    0.97
     weekends
    0.97
    site
    0.96
    boarding
    0.95
    coming
    0.95
    sight
    0.91
    Activations Density 0.233%

    No Known Activations