© 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 12288
    9. 8-RES_FS12288-JB
    11. 1672
    Prev
    Next
    INDEX
    Explanations

    mentions of legal or criminal activities

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    386
    +0.14
    0.6%
    628
    +0.12
    0.5%
    451
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    628
    +0.14
    0.08
    386
    +0.12
    0.09
    451
    +0.10
    0.07
    Negative Logits
    natureconservancy
    -0.76
     Generations
    -0.75
    optim
    -0.72
    growth
    -0.71
    angular
    -0.71
     Wonders
    -0.71
    rients
    -0.71
     optim
    -0.71
     nostalg
    -0.68
     glide
    -0.67
    POSITIVE LOGITS
     misdem
    1.65
     criminally
    1.63
     punishable
    1.61
     felony
    1.59
     misdemeanor
    1.58
     sentenced
    1.58
     jail
    1.58
     jailed
    1.55
     convicted
    1.53
     criminal
    1.52
    Activations Density 1.462%

    No Known Activations