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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 16133
    Prev
    Next
    INDEX
    Explanations

    occurrences of version control comments in code documentation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Beginning of Javadoc-style comments

    np_acts-logits-general · gemini-2.0-flash

    JavaDoc @author

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    +#+#
    -0.59
    сылкі
    -0.56
    tanleria
    -0.54
     Мексичка
    -0.49
    __":
    
    -0.49
    олові
    -0.45
     kaarangay
    -0.44
     violencia
    -0.44
     AppCompatTheme
    -0.44
    AndEndTag
    -0.44
    POSITIVE LOGITS
    author
    0.50
     */
    0.46
    Author
    0.45
     Copyright
    0.44
     author
    0.44
     Author
    0.44
    brev
    0.44
    VersionUID
    0.42
    Created
    0.42
     created
    0.41
    Activations Density 0.044%

    No Known Activations