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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 13144
    Prev
    Next
    INDEX
    Explanations

    phrases related to problem-solving and resolutions

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

    "Solve" or variations of "solve"

    np_acts-logits-general · gemini-2.0-flash

    solve problems

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    ftime
    -0.43
    componentWill
    -0.43
    startAnimation
    -0.42
     profiler
    -0.40
    BindingSource
    -0.39
    ondheim
    -0.38
     semin
    -0.37
     pavilion
    -0.37
     bran
    -0.36
    rawDesc
    -0.36
    POSITIVE LOGITS
     problems
    0.91
    problems
    0.87
     solve
    0.87
    解決
    0.86
     solved
    0.86
     problemen
    0.85
     solving
    0.85
    problemas
    0.84
     solves
    0.82
     problème
    0.81
    Activations Density 0.388%

    No Known Activations