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. 7786
    Prev
    Next
    INDEX
    Explanations

    references to divine authority and prophecy

    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
    +0.25
    0.8%
    453
    +0.12
    0.4%
    1343
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1446
    +0.25
    0.06
    337
    +0.12
    0.04
    185
    +0.10
    0.05
    Negative Logits
    <bos>
    -2.10
    الحياه
    -0.78
    ,
    -0.77
    public
    -0.74
     or
    -0.73
     in
    -0.72
    Tē
    -0.71
    /*
    -0.71
     am
    -0.70
    ,
    -0.70
    POSITIVE LOGITS
     affor
    2.14
     volunte
    2.06
     accla
    2.03
     michelin
    1.97
     reluct
    1.97
     sappi
    1.92
     swarovski
    1.91
     increa
    1.88
     philanth
    1.87
     wherea
    1.87
    Activations Density 0.208%

    No Known Activations