Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 10-RES-JB
    6. 931
    Prev
    Next
    INDEX
    Explanations

    the conjunction "but" and its usage in contrastive contexts

    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.10.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    chanind/openwebtext-gemma
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    1,024
    Dataset
    ctigges/openwebtext-gemma-1024-cl
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.30
    1.3%
    381
    +0.12
    0.5%
    1276
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1276
    +0.30
    0.13
    1068
    +0.12
    0.10
    555
    +0.11
    0.10
    Negative Logits
    <bos>
    -2.97
    expandindo
    -0.79
    endwhile
    -0.75
    ⓧ
    -0.69
    /***
    
    -0.67
    ContentAlignment
    -0.67
    protected
    -0.63
    Enllaços
    -0.61
    }]);
    -0.61
    //{
    
    -0.61
    POSITIVE LOGITS
     impra
    1.62
     reluct
    1.54
     shenan
    1.53
     disagre
    1.49
     hairc
    1.48
     unspeak
    1.48
     impractica
    1.46
     maneu
    1.45
     scrat
    1.45
     swarovski
    1.42
    Activations Density 0.358%

    No Known Activations