© 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. 1220
    Prev
    Next
    INDEX
    Explanations

    phrases expressing worry or anxiety about daily life and its uncertainties

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

    Follows certain prepositions and articles

    np_acts-logits-general · gemini-2.0-flash

    computational programming logic

    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
     <<<<<<<<<<<<<<
    -0.67
     corporativa
    -0.58
     canst
    -0.57
     shalt
    -0.56
     asunder
    -0.56
     Combien
    -0.56
     Jehová
    -0.54
    AddField
    -0.53
     bağlantılar
    -0.53
    Chham
    -0.52
    POSITIVE LOGITS
     }}"></
    0.53
     doInBackground
    0.51
     Sime
    0.47
    "}";
    0.46
    databind
    0.46
    ętr
    0.46
    ArgumentParser
    0.46
    eleste
    0.45
    Ộ
    0.45
    }}">
    0.45
    Activations Density 0.034%

    No Known Activations