© 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
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 7126
    Prev
    Next
    INDEX
    Explanations

    phrases related to conditions and exceptions in procedural contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    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
     hâte
    -0.30
     artificial
    -0.29
     hjelpe
    -0.28
     Uncle
    -0.28
    之间的
    -0.27
    Uncle
    -0.26
     Artificial
    -0.26
     útil
    -0.26
     Natural
    -0.26
    COMPAT
    -0.26
    POSITIVE LOGITS
    ſelben
    0.69
    tanleria
    0.67
     ſtand
    0.67
    ſchaft
    0.64
    iſen
    0.63
     emailAlready
    0.62
     dieſe
    0.62
     myſelf
    0.61
    enablog
    0.61
    ſelf
    0.60
    Activations Density 0.284%

    No Known Activations