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. 16362
    Prev
    Next
    INDEX
    Explanations

    punctuation and sentence structure

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

    A period that marks the end of a paragraph before a transition to a new section, topic heading, or list in structured text.

    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
     tartalomajánló
    -0.41
     препратки
    -0.39
    ulage
    -0.36
    ✭✭
    -0.35
     Оно
    -0.34
     переписи
    -0.34
    zzleHttp
    -0.34
     незавершена
    -0.33
    voice
    -0.33
     Cre
    -0.33
    POSITIVE LOGITS
     waaronder
    0.68
     including
    0.66
     Including
    0.61
     כולל
    0.59
     включая
    0.57
    Например
    0.57
    SharedDtor
    0.56
    Including
    0.55
    including
    0.54
     darunter
    0.54
    Activations Density 0.083%

    No Known Activations