© 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. 2-GEMMASCOPE-RES-16K
    6. 3908
    Prev
    Next
    INDEX
    Explanations

    terms related to passages and connections in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    CloseOperation
    -0.57
     endblock
    -0.49
    endsection
    -0.46
     eşofman
    -0.44
    licants
    -0.43
     costumi
    -0.43
     fals
    -0.41
    তথ্যসূত্র
    -0.41
    Personensuche
    -0.41
    setCharacter
    -0.40
    POSITIVE LOGITS
     Divine
    0.61
     tap
    0.59
     Ale
    0.58
     divine
    0.58
     passage
    0.58
     breakdown
    0.57
    ParallelGroup
    0.56
     &___
    0.56
     decode
    0.56
    Nap
    0.54
    Activations Density 0.505%

    No Known Activations