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 3072
    5. 8-RES_FS3072-JB
    6. 2119
    Prev
    Next
    INDEX
    Explanations

    special characters used in a specific context

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    640
    +0.19
    0.9%
    102
    +0.19
    0.9%
    314
    +0.15
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.19
    0.13
    152
    +0.19
    0.11
    537
    +0.15
    0.10
    Negative Logits
     Led
    -0.82
     Cassidy
    -0.79
     Dickinson
    -0.74
     Barron
    -0.70
    ulhu
    -0.68
     Loll
    -0.68
     Reloaded
    -0.68
     WD
    -0.68
     Rivals
    -0.66
     Kyl
    -0.66
    POSITIVE LOGITS
    existent
    1.42
    government
    1.32
    special
    1.23
    division
    1.23
    exclusive
    1.20
    dist
    1.20
    national
    1.19
    conf
    1.19
    sent
    1.18
    cont
    1.18
    Activations Density 0.416%

    No Known Activations