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 49152
    5. 8-RES_FS49152-JB
    6. 45881
    Prev
    Next
    INDEX
    Explanations

    instances of contrastive conjunctions and qualifiers in sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    393
    +0.10
    0.5%
    732
    +0.10
    0.5%
    95
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.10
    0.05
    271
    +0.10
    0.04
    8
    +0.10
    0.04
    Negative Logits
     Contains
    -0.67
    PDF
    -0.65
    aples
    -0.61
     ®
    -0.60
    Introduction
    -0.60
    Requires
    -0.60
    Reviewed
    -0.59
    arist
    -0.56
     lur
    -0.56
    aily
    -0.54
    POSITIVE LOGITS
     secondly
    1.18
     yeah
    1.14
    entimes
    0.97
     fortunately
    0.96
     luckily
    0.95
     I
    0.95
     unfortunately
    0.95
     everybody
    0.94
     frankly
    0.92
     somebody
    0.90
    Activations Density 0.135%

    No Known Activations