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. 1750
    Prev
    Next
    INDEX
    Explanations

    instances of the word "second" and related terms like "second chance" or "second team"

    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
    373
    +0.10
    0.4%
    296
    +0.09
    0.4%
    582
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    402
    +0.10
    0.07
    701
    +0.09
    0.06
    366
    +0.09
    0.06
    Negative Logits
    Cache
    -0.64
    Statistics
    -0.64
    IPS
    -0.64
    ãĤĭ
    -0.64
    Rules
    -0.64
    cks
    -0.63
    immers
    -0.62
     SPACE
    -0.62
    IRC
    -0.61
    velt
    -0.60
    POSITIVE LOGITS
     baseman
    1.30
    arily
    1.05
    hand
    1.03
     consecutive
    0.91
     cousin
    0.89
     tier
    0.87
     harmonic
    0.86
     successive
    0.86
     inning
    0.85
     installment
    0.85
    Activations Density 0.276%

    No Known Activations