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. 15497
    Prev
    Next
    INDEX
    Explanations

    references to research methodology and publication processes

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

    Academic publications and databases

    np_acts-logits-general · gemini-2.0-flash

    academic journal publisher

    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
     devriez
    -0.43
     soldadura
    -0.42
    :+:
    -0.38
     pauvreté
    -0.37
    nóstico
    -0.37
    #+#
    -0.36
     kyl
    -0.35
     Tuner
    -0.35
    vably
    -0.35
    BeginInit
    -0.35
    POSITIVE LOGITS
     journals
    1.03
     journal
    1.00
     Journals
    0.95
    journals
    0.87
     Journal
    0.84
     JOUR
    0.79
    journal
    0.73
     Scopus
    0.73
    Journal
    0.73
     publisher
    0.72
    Activations Density 2.179%

    No Known Activations