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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 57291
    Prev
    Next
    INDEX
    Explanations

    expressions of personal assurance and recommendation

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

    First-person pronouns followed by verbs that express assurance, communication, or promises in direct address to readers in informal online writing.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    verticalLayout
    -0.38
     Komunikasi
    -0.37
     cuándo
    -0.36
    哪裡
    -0.34
     Wikiseite
    -0.34
     hòa
    -0.33
     Putra
    -0.33
     haka
    -0.33
     cuánt
    -0.33
    哪里
    -0.33
    POSITIVE LOGITS
    rungsseite
    0.60
     незавершена
    0.58
    GTCX
    0.57
     propOrder
    0.52
    ьаж
    0.49
    Personensuche
    0.47
     khuyên
    0.47
    Autoritní
    0.47
     EconPapers
    0.45
    oredCriteria
    0.44
    Activations Density 0.027%

    No Known Activations