© 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. 18575
    Prev
    Next
    INDEX
    Explanations

    countries, cities, or famous names with significant resonance

    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
    373
    +0.28
    1.6%
    232
    +0.12
    0.6%
    750
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.28
    -0.01
    232
    +0.12
    0.08
    406
    +0.09
    0.06
    Negative Logits
    FORMATION
    -0.58
     disguise
    -0.56
     dedicated
    -0.56
     depreciation
    -0.56
     dracon
    -0.55
     partName
    -0.54
     doom
    -0.54
     priceless
    -0.54
     ply
    -0.53
    BILITY
    -0.53
    POSITIVE LOGITS
    wana
    0.85
    ima
    0.83
    uz
    0.83
    amon
    0.81
    aj
    0.81
    ulla
    0.80
    amo
    0.79
    iane
    0.79
    jen
    0.77
    ara
    0.77
    Activations Density 0.206%

    No Known Activations