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. 16399
    Prev
    Next
    INDEX
    Explanations

    discussions of complex social structures and their implications

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

    Common nouns that serve as key concepts within their respective text domains, often representing objects, entities, states, or relationships central to the surrounding context.

    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
    󠁢
    -0.49
     overall
    -0.45
    Overall
    -0.44
    PYX
    -0.44
     Overall
    -0.43
    zehnten
    -0.42
    overall
    -0.42
     nakalista
    -0.42
    medriver
    -0.41
     Italijanski
    -0.39
    POSITIVE LOGITS
     EClass
    0.52
    reathable
    0.47
    LookAnd
    0.45
    +:+
    0.45
     migrants
    0.42
     oxidation
    0.40
    Toponymie
    0.40
     Direct
    0.40
     softly
    0.39
     odkazy
    0.39
    Activations Density 0.077%

    No Known Activations