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. 15750
    Prev
    Next
    INDEX
    Explanations

    references to updates and obligations regarding information accuracy or currency

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

    associated with updating information

    np_acts-logits-general · gemini-2.0-flash

    information currency

    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
    TagMode
    -0.47
    EDEFAULT
    -0.37
     Exactos
    -0.35
     كومونز
    -0.35
     оригіналу
    -0.34
    HomeAsUpEnabled
    -0.34
    发明
    -0.34
    StoryboardSegue
    -0.33
     cherchés
    -0.33
     NSCoder
    -0.32
    POSITIVE LOGITS
     update
    2.59
     updates
    2.39
     updating
    2.38
     updated
    2.30
     Update
    2.23
    update
    2.16
    更新
    2.16
     Updating
    2.13
     Updates
    2.11
    Update
    2.05
    Activations Density 0.317%

    No Known Activations