© 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. 8668
    Prev
    Next
    INDEX
    Explanations

    phrases related to moral values and ethical principles

    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
    481
    +0.15
    0.7%
    507
    +0.13
    0.6%
    447
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    705
    +0.15
    0.05
    601
    +0.13
    0.04
    228
    +0.13
    0.05
    Negative Logits
    backer
    -0.76
    stal
    -0.70
    anie
    -0.70
    ETA
    -0.68
    biz
    -0.66
    Guard
    -0.65
    bg
    -0.63
    inar
    -0.61
    aukee
    -0.61
    gallery
    -0.60
    POSITIVE LOGITS
     there
    0.79
     homosexuality
    0.76
     "[
    0.75
     although
    0.73
     "...
    0.70
     preserving
    0.69
     '[
    0.69
     legalizing
    0.68
     "âĢ¦
    0.67
     they
    0.66
    Activations Density 0.181%

    No Known Activations