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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    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. 45440
    Prev
    Next
    INDEX
    Explanations

    programming and coding-related syntax and keywords

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    arraycopy
    -0.56
     GenerationType
    -0.55
     autorytatywna
    -0.51
    
    -0.50
     Audiodateien
    -0.48
    PDOException
    -0.47
     lateinit
    -0.46
     createSprite
    -0.46
     ſch
    -0.45
    GEBURTS
    -0.45
    POSITIVE LOGITS
    ExtendWith
    0.38
    expandindo
    0.36
    RunWith
    0.35
     Runde
    0.32
     insuffisamment
    0.32
    elemField
    0.31
    ProcessEvent
    0.31
     gius
    0.31
     Summe
    0.31
    ctomy
    0.30
    Activations Density 0.330%

    No Known Activations