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. 1741
    Prev
    Next
    INDEX
    Explanations

    phrases related to problem-solving and proposed solutions

    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
    195
    +0.11
    0.5%
    308
    +0.10
    0.5%
    379
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    195
    +0.11
    0.03
    744
    +0.10
    0.05
    455
    +0.10
    0.06
    Negative Logits
    icipated
    -0.78
     skelet
    -0.73
    ammy
    -0.72
    aquin
    -0.69
    aws
    -0.66
    akin
    -0.64
    avin
    -0.63
    ardy
    -0.61
     Vengeance
    -0.61
     suscept
    -0.61
    POSITIVE LOGITS
    vable
    0.90
     solve
    0.86
    idian
    0.84
    ĵĺ
    0.79
    olution
    0.74
    Problem
    0.71
    ysis
    0.71
    itaire
    0.69
    pton
    0.69
     puzzles
    0.68
    Activations Density 0.829%

    No Known Activations