Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 15594
    Prev
    Next
    INDEX
    Explanations

    words and references related to "Terra" or "Terrestrial" contexts, especially in names and scientific classifications

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

    "Ter" and "Terry"

    np_acts-logits-general · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     createSlice
    -0.50
    argout
    -0.46
    IUrlHelper
    -0.45
     Guess
    -0.43
    StackNavigator
    -0.42
     Grecs
    -0.42
    openzeppelin
    -0.42
    ༘
    -0.42
     Romains
    -0.40
    canestro
    -0.40
    POSITIVE LOGITS
     Ter
    0.65
    Ter
    0.60
     ter
    0.60
     TER
    0.54
     TERR
    0.53
     Terr
    0.51
     Teri
    0.47
     terci
    0.46
     Teres
    0.43
     Terrier
    0.42
    Activations Density 0.154%

    No Known Activations