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 - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 82177
    Prev
    Next
    INDEX
    Explanations

    terms related to the concept of cancellation or errors in communication

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

    cancel, canceled, cancellation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     desafíos
    -0.40
     relève
    -0.40
     suficientemente
    -0.38
     retos
    -0.38
     logros
    -0.37
     défis
    -0.37
    Rohy
    -0.36
     obtend
    -0.36
    twimg
    -0.36
    볼
    -0.36
    POSITIVE LOGITS
     cancelled
    1.21
     cancellation
    1.20
     cancelling
    1.20
     canceled
    1.18
     canceling
    1.16
     Cancelled
    1.13
    Cancelled
    1.13
    cancelled
    1.12
     cancellations
    1.11
     cancel
    1.10
    Activations Density 0.028%

    No Known Activations