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. 1754
    Prev
    Next
    INDEX
    Explanations

    important and serious discussions related to societal, legal, and political matters

    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
    542
    +0.11
    0.5%
    312
    +0.10
    0.5%
    377
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    542
    +0.11
    0.10
    247
    +0.10
    0.08
    312
    +0.10
    0.07
    Negative Logits
    ancies
    -0.92
    utenberg
    -0.78
    assies
    -0.73
    onyms
    -0.73
     chambers
    -0.69
    abella
    -0.69
     roofs
    -0.68
    undreds
    -0.68
    obos
    -0.68
     Regions
    -0.67
    POSITIVE LOGITS
     unto
    1.02
     indeed
    0.92
     nonetheless
    0.90
     akin
    0.89
     worth
    0.88
     breaker
    0.87
     worthy
    0.84
     considering
    0.84
     compared
    0.79
    ifier
    0.79
    Activations Density 1.322%

    No Known Activations