Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 21892
    Prev
    Next
    INDEX
    Explanations

    expressions related to authority, decision-making, and responsibility

    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
    259
    +0.13
    0.6%
    510
    +0.10
    0.5%
    565
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    505
    +0.13
    0.06
    250
    +0.10
    0.04
    259
    +0.10
    0.04
    Negative Logits
    onyms
    -0.82
    idered
    -0.75
     acquaintances
    -0.74
    types
    -0.72
    erald
    -0.69
    osponsors
    -0.69
    cia
    -0.67
    rentices
    -0.67
    olid
    -0.67
    iety
    -0.66
    POSITIVE LOGITS
     overseeing
    1.13
     brunt
    1.04
     responsibility
    0.96
     respons
    0.90
     superv
    0.86
     oversee
    0.82
     reins
    0.82
     orchestr
    0.82
    orship
    0.82
     approving
    0.79
    Activations Density 0.733%

    No Known Activations