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 768
    5. 8-RES_FS768-JB
    6. 158
    Prev
    Next
    INDEX
    Explanations

    technical terms related to legal and intellectual property issues

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs768-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    768
    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
    314
    +0.19
    0.9%
    640
    +0.19
    0.9%
    102
    +0.17
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    152
    +0.19
    0.21
    102
    +0.19
    0.19
    314
    +0.17
    0.05
    Negative Logits
    ulhu
    -0.76
     Dickinson
    -0.73
     Meadows
    -0.72
     Phi
    -0.72
     Mermaid
    -0.71
     Cah
    -0.70
     Gillespie
    -0.69
     Richards
    -0.69
     Reed
    -0.69
     Cassidy
    -0.69
    POSITIVE LOGITS
    sized
    1.40
    level
    1.34
    numbered
    1.28
    government
    1.26
    interest
    1.25
    trial
    1.24
    generation
    1.24
    division
    1.23
    circ
    1.23
    serving
    1.23
    Activations Density 1.220%

    No Known Activations