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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 6126
    Prev
    Next
    INDEX
    Explanations

    phrases related to caution and decision-making in political contexts

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

    Preceding "is," "to," or "get."

    np_acts-logits-general · gemini-2.0-flash

    underestimating mistake caught following dwelling

    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
     honte
    -0.66
    íticas
    -0.52
    delwed
    -0.50
     pijn
    -0.50
    Portály
    -0.49
    ítica
    -0.49
     hated
    -0.49
    恶心
    -0.48
    Weird
    -0.47
     silêncio
    -0.47
    POSITIVE LOGITS
     hasty
    1.01
     misguided
    1.00
     unwise
    0.97
     imprud
    0.94
     naive
    0.93
     mistakenly
    0.93
     nai
    0.93
     blindly
    0.91
     rash
    0.90
     mistaken
    0.90
    Activations Density 0.476%

    No Known Activations