Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 14536
    Prev
    Next
    INDEX
    Explanations

    elements related to figures and data representations in scientific research

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

    Filenames in image references

    np_acts-logits-general · gemini-2.0-flash

    Ann Gastroenterol

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     telefónica
    -0.60
     grabación
    -0.48
     hâte
    -0.44
     rumahnya
    -0.43
     hormigón
    -0.43
     superfície
    -0.43
     nécessairement
    -0.42
     portée
    -0.41
     permukaan
    -0.41
     ladrillo
    -0.41
    POSITIVE LOGITS
     linkovi
    0.60
    setVerticalGroup
    0.51
    __*/
    0.50
    __(/*!
    0.50
     Савезне
    0.50
     jadx
    0.47
     poster
    0.45
    الدراسه
    0.45
     Glide
    0.45
     Blitz
    0.44
    Activations Density 0.340%

    No Known Activations