© 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-IT
    4. Residual Stream - 16k
    5. 9-GEMMASCOPE-RES-16K
    6. 10871
    Prev
    Next
    INDEX
    Explanations

    terms related to unexpected outcomes or victories in competitive contexts

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

    Unexpected victory or loss

    np_acts-logits-general · gemini-2.0-flash

    competition outcomes

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    GraphicsUnit
    -0.49
    Enllaces
    -0.46
    Personendaten
    -0.44
     paran
    -0.44
     跳转至
    -0.43
    puta
    -0.42
    
    -0.42
    Personensuche
    -0.41
    :✨
    -0.40
    $__
    -0.40
    POSITIVE LOGITS
     upset
    0.65
     upsets
    0.59
     underdog
    0.51
     upsetting
    0.47
     defeat
    0.47
     defeating
    0.47
     defeated
    0.45
    impresa
    0.42
     eliminación
    0.42
     derrot
    0.42
    Activations Density 0.358%

    No Known Activations