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 · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 12-RES-JB
    6. 182
    Prev
    Next
    INDEX
    Explanations

    quantifiable statistics or proportions related to populations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.12.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    12
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.33
    1.1%
    1252
    +0.09
    0.3%
    198
    +0.07
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1252
    +0.33
    0.04
    1726
    +0.09
    0.03
    125
    +0.07
    0.03
    Negative Logits
    <bos>
    -1.80
     utop
    -0.99
     anse
    -0.94
     solidar
    -0.94
     meras
    -0.92
     sò
    -0.92
     robus
    -0.91
     ù
    -0.90
     rè
    -0.86
     tà
    -0.84
    POSITIVE LOGITS
     of
    0.96
     Of
    0.69
     ofthe
    0.67
     soulign
    0.66
     Putih
    0.66
    ของ
    0.65
     của
    0.64
     doté
    0.63
     fabricado
    0.61
     unwarran
    0.61
    Activations Density 0.121%

    No Known Activations