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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 56434
    Prev
    Next
    INDEX
    Explanations

    mentions of audio and sound-related terminology

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

    audio engineering and recordings

    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
    interess
    -0.47
    inqui
    -0.42
    telling
    -0.40
     inhabitant
    -0.40
    nth
    -0.39
    JKLM
    -0.39
    return
    -0.39
    where
    -0.38
    using
    -0.38
    with
    -0.38
    POSITIVE LOGITS
    Audio
    1.20
     Audio
    1.14
     audio
    1.09
     AUDIO
    1.05
    audio
    1.02
    AUDIO
    0.91
    udio
    0.72
     áudio
    0.71
     Radio
    0.70
     Auditing
    0.70
    Activations Density 0.007%

    No Known Activations