© 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-2-9B
    4. Residual Stream - 16k
    5. 38-GEMMASCOPE-RES-16K
    6. 7959
    Prev
    Next
    INDEX
    Explanations

    references to the term "Mat" and its variations, indicating a focus on mathematical or computational contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aarrggbb
    -0.60
     noDo
    -0.57
    帖最后由
    -0.56
     autorytatywna
    -0.54
    Werdegang
    -0.51
    tagHelperRunner
    -0.50
    出版年
    -0.49
     nahilalakip
    -0.48
    ckså
    -0.46
    Искәрмәләр
    -0.45
    POSITIVE LOGITS
    anyahu
    0.45
     postIndex
    0.42
     fact
    0.40
     defStyleAttr
    0.40
    éo
    0.38
    plotlib
    0.37
     Damon
    0.36
     Ressources
    0.36
    Suara
    0.36
    urities
    0.35
    Activations Density 0.215%

    No Known Activations