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 1536
    5. 8-RES_FS1536-JB
    6. 417
    Prev
    Next
    INDEX
    Explanations

    terms related to voting, ballots, and election processes

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    496
    +0.14
    0.6%
    540
    +0.10
    0.5%
    339
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.10
    540
    +0.10
    0.09
    531
    +0.10
    0.08
    Negative Logits
     Flavoring
    -0.94
     Palestin
    -0.70
    ienne
    -0.70
    ĸļ
    -0.68
    cer
    -0.68
    ofi
    -0.68
     Emer
    -0.67
     Kut
    -0.64
    ModLoader
    -0.64
    Description
    -0.64
    POSITIVE LOGITS
    eering
    1.02
     ballots
    0.99
     votes
    0.97
    ndum
    0.94
     voting
    0.94
     vote
    0.92
     polls
    0.92
    vote
    0.87
     booth
    0.86
     turnout
    0.86
    Activations Density 1.654%

    No Known Activations