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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 28-GEMMASCOPE-RES-16K
    6. 8015
    Prev
    Next
    INDEX
    Explanations

    directives related to safety precautions and requirements

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

    suggesting caution or warning

    np_acts-logits-general · gemini-2.0-flash

    be careful

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.28.hook_resid_post
    Hook Layer
    28
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    writeField
    -0.35
    
    -0.34
     doInBackground
    -0.31
    Hochspringen
    -0.30
     setShow
    -0.30
     poveznice
    -0.30
    lebnis
    -0.29
    balleur
    -0.29
    wüns
    -0.29
     continúas
    -0.28
    POSITIVE LOGITS
     caution
    0.98
     careful
    0.92
     cautious
    0.92
     precautions
    0.90
     precaution
    0.83
     Careful
    0.80
     carefully
    0.79
    caution
    0.79
     Caution
    0.79
    Careful
    0.78
    Activations Density 0.447%

    No Known Activations