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. 1534
    Prev
    Next
    INDEX
    Explanations

    keywords related to programming functions and structures

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

    ending in "ed"

    np_acts-logits-general · gemini-2.0-flash

    past tense / participial verbs

    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
     houſe
    -0.72
     pleaſure
    -0.66
     Houſe
    -0.59
     ſche
    -0.59
     itſelf
    -0.59
     purpoſe
    -0.58
     ſch
    -0.58
    ſelf
    -0.57
     juges
    -0.56
     perſon
    -0.56
    POSITIVE LOGITS
    ized
    0.77
    ourced
    0.77
    ted
    0.76
    ded
    0.75
    lared
    0.74
    aced
    0.73
    iled
    0.72
    iked
    0.71
    tered
    0.71
    ched
    0.71
    Activations Density 1.095%

    No Known Activations