Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 3176
    Prev
    Next
    INDEX
    Explanations

    aspects of healthcare and medical conditions

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

    Text after questions marked with "Q:"

    np_acts-logits-general · gemini-2.0-flash

    introductions and questions

    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
    tvguidetime
    -0.90
    Czytaj
    -0.84
     BoxDecoration
    -0.76
     henvisninger
    -0.75
    abcdefghijklmnop
    -0.74
     发表于
    -0.74
     CreateTagHelper
    -0.73
    曖昧さ回避
    -0.70
     intptr
    -0.68
     Numerade
    -0.67
    POSITIVE LOGITS
    “
    0.66
    bcryptjs
    0.65
    󠁿
    0.62
    تیب
    0.58
    parsedMessage
    0.57
    ‘
    0.50
    StructEnd
    0.49
    Rourke
    0.48
    ing
    0.47
    ”
    0.46
    Activations Density 0.870%

    No Known Activations