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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 92902
    Prev
    Next
    INDEX
    Explanations

    phrases indicating offers of assistance or support

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

    offering or requesting help

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     crash
    -0.40
     performer
    -0.38
     bolt
    -0.38
    didSet
    -0.37
    ributed
    -0.35
     bulletin
    -0.35
    広告
    -0.34
    ặt
    -0.34
     Privile
    -0.34
     favoured
    -0.34
    POSITIVE LOGITS
    printStackTrace
    0.60
     Efq
    0.54
     Diſ
    0.54
     EconPapers
    0.54
     segíts
    0.53
     canst
    0.52
     Majefty
    0.52
    AndroidJUnit
    0.52
     Perſ
    0.51
    AnchorStyles
    0.51
    Activations Density 0.008%

    No Known Activations