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 6144
    5. 8-RES_FS6144-JB
    6. 5354
    Prev
    Next
    INDEX
    Explanations

    sentences excluding specific contrasting exceptions indicated by the word "except"

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs6144-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    6,144
    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
    377
    +0.10
    0.5%
    373
    +0.10
    0.5%
    275
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    747
    +0.10
    0.03
    8
    +0.10
    0.03
    33
    +0.10
    0.03
    Negative Logits
    natureconservancy
    -0.69
    ierce
    -0.63
     enthusi
    -0.61
    eny
    -0.60
    venture
    -0.59
     acceler
    -0.58
    piring
    -0.55
     redevelop
    -0.55
     masculinity
    -0.55
    iable
    -0.55
    POSITIVE LOGITS
    ional
    1.22
     insofar
    0.90
     maybe
    0.86
     perhaps
    0.75
     occasional
    0.73
     possibly
    0.68
    sie
    0.64
    ments
    0.63
    icut
    0.63
    ords
    0.63
    Activations Density 0.215%

    No Known Activations