© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 22-GEMMASCOPE-RES-16K
    11. 3541
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions and logical constructs in formal language

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

    Mathematical or logical formulas

    np_acts-logits-general · gemini-2.0-flash

    logic and computation

    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
    #+#
    -0.64
    AnchorStyles
    -0.53
    Péri
    -0.50
     Jost
    -0.50
     Biographical
    -0.49
     romanti
    -0.48
    BeginInit
    -0.48
     actionTypes
    -0.48
     skid
    -0.48
    ContentAsync
    -0.48
    POSITIVE LOGITS
     ¬
    0.68
     clauses
    0.61
     proposition
    0.60
     clause
    0.57
     skole
    0.57
    Personendaten
    0.55
     truth
    0.55
     proof
    0.53
     grounding
    0.52
    Fresh
    0.52
    Activations Density 1.037%

    No Known Activations