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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 10385
    Prev
    Next
    INDEX
    Explanations

    references to historical events involving battles and their effects

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

    Occurrences after "at the Battle of"

    np_acts-logits-general · gemini-2.0-flash

    battles and defeats

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    gwt
    -0.42
    alina
    -0.42
     AspNetCore
    -0.40
    kunt
    -0.40
    enschein
    -0.39
     protestas
    -0.39
    ья
    -0.38
    leuth
    -0.38
     /*
    -0.37
    yelidikan
    -0.37
    POSITIVE LOGITS
     defeat
    1.10
     defeating
    0.99
     battle
    0.97
     defeats
    0.96
     battles
    0.96
     recapture
    0.95
    Battle
    0.95
     Defeat
    0.95
     Battle
    0.94
     nakalista
    0.94
    Activations Density 0.461%

    No Known Activations