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. 1155
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal proceedings and their nuances

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    a contrast or opposition

    np_acts-logits-general · gemini-2.0-flash

    success and failure outcomes

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    oglobulin
    -0.49
     Intensity
    -0.48
     Compens
    -0.47
    tonode
    -0.46
    исленность
    -0.46
     AssemblyVersion
    -0.45
    laston
    -0.45
    Distribuzione
    -0.44
     USM
    -0.44
    EPI
    -0.43
    POSITIVE LOGITS
    tvguidetime
    0.50
    ⓧ
    0.39
    MarshalTo
    0.39
    ("")]
    
    0.38
    astify
    0.37
     faſt
    0.36
     rather
    0.35
     quæ
    0.35
    ]")]
    0.35
    Tikang
    0.35
    Activations Density 0.983%

    No Known Activations