© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 105007
    Prev
    Next
    INDEX
    Explanations

    references to steepness or incline in various contexts

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

    steepness and steep gradients

    np_acts-logits-general · gemini-2.5-flash-lite

    references to steepness—highlighting mentions of steep slopes, ascents, gradients, or rates, especially in geographic or quantitative contexts.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    UserScript
    -0.72
    ngrx
    -0.63
    astify
    -0.60
     doInBackground
    -0.60
     Normdatei
    -0.60
     informée
    -0.60
    userManager
    -0.59
     Colonies
    -0.59
     Cosmos
    -0.59
     ControllerBase
    -0.59
    POSITIVE LOGITS
     Steep
    0.95
     steep
    0.85
     steeper
    0.62
     steeply
    0.61
    陡
    0.55
     ſte
    0.46
     steeple
    0.45
    Deep
    0.42
     escarp
    0.42
     Deep
    0.41
    Activations Density 0.004%

    No Known Activations