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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 65026
    Prev
    Next
    INDEX
    Explanations

    keywords and phrases about local food culture and experiences

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

    not so, claim to, sucker for

    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-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
    principalColumn
    -0.63
    MessageTagHelper
    -0.60
     ब्रेकडाउन
    -0.58
     esternos
    -0.54
    BeginContext
    -0.53
    
    -0.53
    piram
    -0.44
    szönöm
    -0.43
    发表于
    -0.41
    therefore
    -0.41
    POSITIVE LOGITS
     gł
    0.41
    styleType
    0.36
     Liga
    0.36
     friends
    0.34
    findung
    0.34
    받
    0.34
     Oficial
    0.33
    ">—
    0.33
    anera
    0.33
     Bé
    0.33
    Activations Density 0.042%

    No Known Activations