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. 4118
    Prev
    Next
    INDEX
    Explanations

    special characters or unique symbols in the document

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

    German letters (ü, ö, ä)

    np_acts-logits-general · gemini-2.0-flash

    german umlauts and place names

    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
     continúas
    -0.43
    Ārējās
    -0.43
     linguri
    -0.42
     Heu
    -0.41
     Quell
    -0.41
     Lage
    -0.41
     Raub
    -0.40
    ldern
    -0.40
     Heft
    -0.40
     albeit
    -0.38
    POSITIVE LOGITS
     betweenstory
    0.60
     queſta
    0.50
     ***!
    0.50
    AndEndTag
    0.49
     للمعارف
    0.48
    -------------</
    0.48
     ſch
    0.47
     invokingState
    0.47
     ſta
    0.46
     للاسماء
    0.46
    Activations Density 0.052%

    No Known Activations