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
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 14920
    Prev
    Next
    INDEX
    Explanations

    information related to scientific accuracy and measurements in research contexts

    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
    vosť
    -0.41
    UnifiedTopology
    -0.41
     PROFES
    -0.39
     tarko
    -0.39
     känna
    -0.38
    rierte
    -0.37
    HasAnnotation
    -0.36
    CEPTION
    -0.36
     aban
    -0.36
     informée
    -0.36
    POSITIVE LOGITS
    simply
    0.45
    単純
    0.45
     Signalez
    0.43
    
    0.43
     simply
    0.43
    MLLoader
    0.43
     Simply
    0.42
     ordinary
    0.42
     invokingState
    0.42
     semplicemente
    0.41
    Activations Density 1.194%

    No Known Activations