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. 7161
    Prev
    Next
    INDEX
    Explanations

    phrases associated with feeling better or recovery

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

    feeling better or improved states

    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
     plomo
    -0.36
    ंटर
    -0.35
     ultra
    -0.33
    jiny
    -0.33
    oarece
    -0.32
    inheritdoc
    -0.32
    дарь
    -0.32
    之地
    -0.31
     glory
    -0.31
    "}";
    -0.31
    POSITIVE LOGITS
     فريبيس
    0.67
    rungsseite
    0.66
     betweenstory
    0.63
    recovered
    0.58
     متعلقه
    0.57
     Recovery
    0.56
    Recovered
    0.56
    recovery
    0.56
     recovered
    0.56
     recovery
    0.54
    Activations Density 0.008%

    No Known Activations