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. 1496
    Prev
    Next
    INDEX
    Explanations

    specific categories of products and services, particularly in areas such as technology, healthcare, and consumer goods

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

    following specific types of equipment

    np_acts-logits-general · gemini-2.0-flash

    disease, fumes, skating, shows

    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
     du
    -0.33
     kwar
    -0.32
     ergeben
    -0.32
     isolated
    -0.31
     height
    -0.30
     Final
    -0.29
     contract
    -0.28
     ex
    -0.28
     расс
    -0.28
     comp
    -0.27
    POSITIVE LOGITS
    MessageOf
    0.63
     pinulongan
    0.60
    ⤹
    0.58
     ControllerBase
    0.56
     مرئيه
    0.55
    #+#
    0.55
    ьаж
    0.54
    +#+#
    0.54
    SequentialGroup
    0.53
    StoreMessageInfo
    0.53
    Activations Density 0.725%

    No Known Activations