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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 8469
    Prev
    Next
    INDEX
    Explanations

    references related to physical locations and actions associated with them

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     avoient
    -0.44
    !*\
    -0.43
    ])).
    -0.41
     enfans
    -0.39
    تقاوى
    -0.37
     nebst
    -0.37
    ]));
    
    -0.37
    })).
    -0.36
    出版年
    -0.35
     اطلع
    -0.35
    POSITIVE LOGITS
    存于互联网档案馆
    0.77
    withIdentifier
    0.62
     Normdatei
    0.62
    athyroid
    0.60
    databind
    0.59
    SequentialGroup
    0.58
    openConnection
    0.57
    Aiheesta
    0.56
    readObject
    0.53
    zzleHttp
    0.52
    Activations Density 0.446%

    No Known Activations