© 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. 1815
    Prev
    Next
    INDEX
    Explanations

    statements related to legal or political contexts

    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
    7
    +0.12
    0.6%
    134
    +0.12
    0.5%
    720
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.12
    0.10
    55
    +0.12
    0.08
    7
    +0.10
    0.06
    Negative Logits
     Horizons
    -0.75
     Teg
    -0.68
     Arri
    -0.68
     Telescope
    -0.67
    banks
    -0.67
     Finding
    -0.66
     Centauri
    -0.66
     Explorer
    -0.65
     Hosp
    -0.65
     Vall
    -0.65
    POSITIVE LOGITS
     punishable
    1.28
     unlawful
    1.28
     perpetrated
    1.27
     permissible
    1.16
     lawful
    1.16
     prohibited
    1.11
     illegal
    1.06
     immoral
    1.03
     consensual
    1.03
     occurring
    1.03
    Activations Density 0.828%

    No Known Activations