© 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. 6983
    Prev
    Next
    INDEX
    Explanations

    social sharing-related terms

    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
    39
    +0.11
    0.5%
    679
    +0.10
    0.5%
    357
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    39
    +0.11
    0.02
    357
    +0.10
    0.02
    74
    +0.10
    0.02
    Negative Logits
    schild
    -0.82
     destro
    -0.73
    anthrop
    -0.68
    paio
    -0.66
    iatric
    -0.65
    interrupted
    -0.61
    issance
    -0.61
    quarter
    -0.61
    orrow
    -0.60
    wagen
    -0.60
    POSITIVE LOGITS
    holder
    1.24
    holders
    1.23
    Tweet
    0.96
    edin
    0.90
    ership
    0.89
    ables
    0.88
     Share
    0.87
     Tweet
    0.87
    able
    0.84
    holding
    0.82
    Activations Density 0.015%

    No Known Activations