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. 130988
    Prev
    Next
    INDEX
    Explanations

    sections that summarize information or highlight main points

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

    Punctuation marks that introduce explanations (":") or summarize methods ("Briefly,"), and formatting symbols ("--", "<<", ">>") that serve as section dividers or message boundaries in technical documentation and academic papers.

    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
     незавершена
    -0.67
    IsMutable
    -0.51
    NameInMap
    -0.50
     typelib
    -0.47
     kaarangay
    -0.47
    цездатний
    -0.45
    تقاوى
    -0.44
     cadeiras
    -0.43
     conmigo
    -0.42
     prohibido
    -0.41
    POSITIVE LOGITS
    
    0.64
    Kaieteur
    0.53
     Protein
    0.51
    XtraEditors
    0.50
     VIDEOTAPE
    0.47
     Packages
    0.46
     Posted
    0.45
    Standort
    0.45
    XmlSchema
    0.44
    Protein
    0.44
    Activations Density 0.101%

    No Known Activations