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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 20-GEMMASCOPE-RES-16K
    6. 15198
    Prev
    Next
    INDEX
    Explanations

    instances of body language and physical interactions

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

    similarity

    np_acts-logits-general · gemini-2.0-flash

    likewise similarly same ditto

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    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
     مرئيه
    -0.50
    KommentareTeilen
    -0.47
    şiv
    -0.46
     Wiktionnaire
    -0.46
    انيف
    -0.45
    FormTagHelper
    -0.42
     noDo
    -0.42
    ngrx
    -0.42
    GetMapping
    -0.40
     kapturem
    -0.40
    POSITIVE LOGITS
     likewise
    0.56
     ebenfalls
    0.56
     similarly
    0.55
     nakalista
    0.52
    同樣
    0.50
    同样
    0.49
    ditto
    0.45
     same
    0.44
     тоже
    0.44
    こちらも
    0.44
    Activations Density 1.420%

    No Known Activations