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. 33-GEMMASCOPE-RES-16K
    6. 11324
    Prev
    Next
    INDEX
    Explanations

    references to military operations and investigations involving experimentation or evaluations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.33.hook_resid_post
    Hook Layer
    33
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AddTagHelper
    -0.36
     createState
    -0.36
     READY
    -0.34
     sẵn
    -0.33
     ready
    -0.32
     Ready
    -0.32
    istoitu
    -0.31
     linkovi
    -0.31
    ready
    -0.31
     موجود
    -0.31
    POSITIVE LOGITS
     involving
    1.20
     aimed
    1.02
     visant
    0.96
     undertaken
    0.95
     conducted
    0.95
    conducted
    0.86
     targeting
    0.83
     carried
    0.81
     aiming
    0.78
    targeting
    0.76
    Activations Density 1.368%

    No Known Activations