© 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-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 13078
    Prev
    Next
    INDEX
    Explanations

    references to workout preparations and routines, particularly warm-ups and injury prevention techniques

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

    warm-ups and warming

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     projekt
    -0.37
     auroit
    -0.37
    プター
    -0.36
     Fermentation
    -0.35
     chrétien
    -0.35
     colher
    -0.34
    èdia
    -0.33
    issors
    -0.33
    })->
    -0.33
     gober
    -0.32
    POSITIVE LOGITS
     warm
    1.37
     warmup
    1.27
    Warm
    1.23
     Warm
    1.23
    warm
    1.20
     warms
    1.20
     warming
    1.18
     warmed
    1.13
    warming
    1.03
     WARM
    1.03
    Activations Density 0.026%

    No Known Activations