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-2-2B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 8763
    Prev
    Next
    INDEX
    Explanations

    references to minimum values and constraints in data structures or configurations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    
    -0.70
    gri
    -0.70
     Smack
    -0.61
     générale
    -0.60
     financière
    -0.59
     FileWriter
    -0.59
     Jwt
    -0.59
    Pok
    -0.59
     SPH
    -0.58
     byteArray
    -0.58
    POSITIVE LOGITS
     MIN
    1.23
    MIN
    1.21
     Min
    1.09
    getMin
    1.07
    Min
    1.04
     Mins
    1.03
    min
    1.01
    Mins
    0.98
     getMin
    0.97
     minions
    0.93
    Activations Density 0.147%

    No Known Activations