© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 2353
    Prev
    Next
    INDEX
    Explanations

    references to episodes and their details within a series

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_15/width_16k/average_l0_131
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     mView
    -0.57
     tää
    -0.54
    dafx
    -0.54
    extAlignment
    -0.52
     tarko
    -0.51
     InputDecoration
    -0.50
     voici
    -0.49
     FirebaseAuth
    -0.48
     Hoo
    -0.48
    ########.
    -0.47
    POSITIVE LOGITS
     episode
    0.69
     island
    0.68
     Island
    0.67
    Island
    0.63
     islands
    0.61
     episodes
    0.61
    episode
    0.61
     epiz
    0.58
     eiland
    0.58
    Episode
    0.58
    Activations Density 0.181%

    No Known Activations