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-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 8681
    Prev
    Next
    INDEX
    Explanations

    references to personal experiences and suggestions for improvements or changes in practices

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

    URLs and related terms

    np_acts-logits-general · gemini-2.0-flash

    http URLs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Ведь
    -0.61
     ddelweddau
    -0.61
    RegressionTest
    -0.61
    SequentialGroup
    -0.60
    だからこそ
    -0.57
     مرئيه
    -0.56
    انيف
    -0.56
    mybatisplus
    -0.56
    ведь
    -0.55
    Tembelea
    -0.55
    POSITIVE LOGITS
     http
    1.00
    http
    0.92
     IIRC
    0.91
     iirc
    0.88
     https
    0.85
    https
    0.80
     recently
    0.77
    recently
    0.73
     excellent
    0.72
    excellent
    0.72
    Activations Density 0.433%

    No Known Activations