© 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 - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 4788
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate commitment or targeted goals related to actions and intentions

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

    Prepositions followed by gerunds

    np_acts-logits-general · gemini-2.0-flash

    purpose or aim

    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
     jamb
    -0.48
     testCase
    -0.47
    FromNib
    -0.45
     tagName
    -0.43
     AssemblyCulture
    -0.41
     Coconut
    -0.41
     epi
    -0.40
     GetComponent
    -0.40
     fullName
    -0.40
    NSNotification
    -0.40
    POSITIVE LOGITS
     ensure
    0.75
     ensuring
    0.73
     ensures
    0.62
     udrž
    0.60
    确保
    0.59
    Ensure
    0.54
    Gön
    0.54
     garantiza
    0.54
     möglichst
    0.53
     sağlay
    0.53
    Activations Density 0.182%

    No Known Activations