© 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. 18769
    Prev
    Next
    INDEX
    Explanations

    statements expressing honesty or frankness

    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
    200
    +0.11
    0.5%
    0
    +0.10
    0.4%
    312
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    246
    +0.11
    0.02
    312
    +0.10
    0.02
    596
    +0.09
    0.02
    Negative Logits
    etting
    -0.69
     Blades
    -0.68
    arthy
    -0.66
    ammy
    -0.65
    tailed
    -0.65
    ied
    -0.64
     Landing
    -0.64
    tein
    -0.60
     Klu
    -0.59
    lav
    -0.59
    POSITIVE LOGITS
     speaking
    1.04
    é¾įåĸļ士
    0.85
    zers
    0.84
    speaking
    0.75
     honestly
    0.73
    ometry
    0.67
     admit
    0.67
     tho
    0.66
     doubted
    0.65
     ashamed
    0.64
    Activations Density 0.032%

    No Known Activations