© 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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 67775
    Prev
    Next
    INDEX
    Explanations

    mentions of assistance or offering help

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

    help followed by recipient or context

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     succinctly
    -0.47
     primordial
    -0.45
     exquisitely
    -0.44
     prior
    -0.44
     Roca
    -0.44
     extremely
    -0.44
     MacGregor
    -0.43
     continuously
    -0.43
     exactly
    -0.43
     meticulously
    -0.43
    POSITIVE LOGITS
    help
    1.23
    Help
    1.13
     Help
    1.07
    HELP
    1.01
    helped
    0.99
    helping
    0.98
    helps
    0.96
     ayuda
    0.95
     help
    0.94
     HELP
    0.94
    Activations Density 0.025%

    No Known Activations