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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 40101
    Prev
    Next
    INDEX
    Explanations

    references to websites and online platforms

    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
    388
    +0.13
    0.6%
    125
    +0.12
    0.5%
    679
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    388
    +0.13
    0.03
    498
    +0.12
    0.03
    679
    +0.10
    0.04
    Negative Logits
     Lilly
    -0.78
     recoil
    -0.72
     sequencing
    -0.71
    felt
    -0.71
    Grade
    -0.68
     proportions
    -0.67
     reductions
    -0.66
    è¦ļéĨĴ
    -0.66
     flu
    -0.65
     anesthesia
    -0.63
    POSITIVE LOGITS
     homepage
    1.25
    bsite
    1.24
     www
    1.22
    accessible
    1.11
     hosted
    1.09
    pages
    1.08
    Website
    1.07
     URL
    1.03
     website
    1.02
     "/
    1.02
    Activations Density 0.204%

    No Known Activations