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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 30176
    Prev
    Next
    INDEX
    Explanations

    hexadecimal values and constants related to programming

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

    0xff ; quant

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rungsseite
    -0.60
    MessageTagHelper
    -0.56
    ✨:
    -0.50
    Distribuzione
    -0.50
    wachsene
    -0.50
     audiences
    -0.49
     typelib
    -0.48
    Chham
    -0.48
    
    -0.48
    addCriterion
    -0.46
    POSITIVE LOGITS
    كويكب
    0.47
     invokingState
    0.46
     SafeMath
    0.35
     quant
    0.34
     FOREVER
    0.33
    cious
    0.33
     quantization
    0.32
     Atl
    0.31
    Rujuakan
    0.30
     ketat
    0.30
    Activations Density 0.006%

    No Known Activations