© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 66132
    Prev
    Next
    INDEX
    Explanations

    terms related to distinguishing or differentiating between concepts or entities

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

    making distinctions

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
     endfor
    -0.48
    iotensin
    -0.48
    WriteTagHelper
    -0.47
    CompleteListener
    -0.47
     totalSupply
    -0.47
    deus
    -0.45
    Cert
    -0.43
    Footer
    -0.43
     épaules
    -0.42
    crollView
    -0.42
    POSITIVE LOGITS
     distinguish
    0.86
     distinctions
    0.84
     distinction
    0.83
     distinguir
    0.82
    distingu
    0.82
     distinguishing
    0.81
     distinguishes
    0.80
     distinción
    0.78
     differentiate
    0.76
     distingu
    0.74
    Activations Density 0.069%

    No Known Activations