© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 2301
    Prev
    Next
    INDEX
    Explanations

    Japanese words and phrases, particularly those related to names and greetings

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

    Japanese language characters and words, particularly those related to proper names and linguistic details.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    WireFormatLite
    -0.46
    
    -0.38
     незавершена
    -0.38
     Schließlich
    -0.37
    PerformLayout
    -0.37
     Zustimmung
    -0.36
     disambiguazione
    -0.36
     Erwä
    -0.36
     personnalisée
    -0.35
     menengah
    -0.35
    POSITIVE LOGITS
    MessageOf
    0.61
    PreInfinity
    0.56
     emb
    0.50
    dic
    0.47
    Astr
    0.47
     اخت
    0.47
    mergeFrom
    0.46
     colli
    0.46
     lo
    0.46
    hang
    0.45
    Activations Density 0.085%

    No Known Activations