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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 15324
    Prev
    Next
    INDEX
    Explanations

    key information and metrics about various subjects, particularly focusing on usefulness and importance

    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
    彼らは
    -0.37
    OGND
    -0.37
    Rujuakan
    -0.36
    $=\
    -0.34
     třeba
    -0.34
     $('.
    -0.34
     â
    -0.33
    																			
    -0.33
     Mole
    -0.32
    قایناق‌لار
    -0.32
    POSITIVE LOGITS
     surla
    0.70
    +#+
    0.61
     natomiast
    0.59
    CloseOperation
    0.56
    pecabe
    0.55
     Verfügung
    0.54
     Israël
    0.54
     Verhält
    0.53
     Verhältnisse
    0.53
     Gesetze
    0.53
    Activations Density 1.041%

    No Known Activations