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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 108199
    Prev
    Next
    INDEX
    Explanations

    phrases indicating methods or means of achieving something

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

    by means of

    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
     itſelf
    -0.59
     pleaſure
    -0.58
     themſelves
    -0.58
     Theſe
    -0.53
    mær
    -0.51
    ſelves
    -0.51
     întâ
    -0.50
     myſelf
    -0.49
     himſelf
    -0.49
     avoient
    -0.49
    POSITIVE LOGITS
     mittels
    0.72
     using
    0.70
     mediante
    0.68
     tramite
    0.66
     допомогою
    0.66
     USING
    0.63
     بواسطة
    0.63
     利用
    0.62
     通过
    0.62
     utilizing
    0.61
    Activations Density 0.018%

    No Known Activations