© 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. 18-GEMMASCOPE-RES-16K
    6. 8353
    Prev
    Next
    INDEX
    Explanations

    noun phrases related to groups or teams

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

    Followed by a verb

    np_acts-logits-general · gemini-2.0-flash

    mentions "authors" or "researchers"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ,
    -0.85
     because
    -0.65
    自从
    -0.59
    からです
    -0.56
    because
    -0.54
    之所以
    -0.54
     after
    -0.52
    嘛
    -0.51
     після
    -0.51
    dientemente
    -0.49
    POSITIVE LOGITS
     חיצוניים
    0.88
     nonetheless
    0.85
     nevertheless
    0.84
     surely
    0.69
     voici
    0.68
     toch
    0.67
     muſt
    0.65
    theless
    0.64
    etheless
    0.62
    SerializedName
    0.62
    Activations Density 0.676%

    No Known Activations