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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 15561
    Prev
    Next
    INDEX
    Explanations

    discussions related to potential implications and statistics surrounding policies and societal issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tagHelperRunner
    -0.92
     الحره
    -0.83
     ब्रेकडाउन
    -0.79
     utafitiHapana
    -0.76
    __':
    
    -0.75
     مشين
    -0.72
     queſta
    -0.71
     transfieras
    -0.70
     tartalomajánló
    -0.70
    IsContent
    -0.69
    POSITIVE LOGITS
     And
    0.56
    ________________
    0.49
    And
    0.49
    <eos>
    0.48
    ↵
    0.48
     Overall
    0.47
     All
    0.46
    ----------------
    0.46
    ↵↵
    0.46
    .
    0.46
    Activations Density 0.375%

    No Known Activations