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. 122953
    Prev
    Next
    INDEX
    Explanations

    elements related to programming and coding performance

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

    Technical terminology that appears within specialized contexts across various domains including computing, mathematics, legal writing, and scientific documentation.

    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
     CreateTagHelper
    -0.48
    hilangan
    -0.47
     mendatang
    -0.46
     Grüsse
    -0.44
     Producción
    -0.43
     Wikiseite
    -0.43
     Schluß
    -0.41
     bahía
    -0.40
     gesunde
    -0.40
     soeur
    -0.40
    POSITIVE LOGITS
    tagHelperRunner
    0.52
     mone
    0.50
     صوتيه
    0.47
    MessageOf
    0.46
     assimilation
    0.45
     interpretations
    0.44
     RISE
    0.44
     absorption
    0.43
     Pli
    0.43
     maxim
    0.42
    Activations Density 0.230%

    No Known Activations