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. 16-GEMMASCOPE-RES-16K
    6. 91
    Prev
    Next
    INDEX
    Explanations

    personal reflections and opinions on experiences or preferences

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

    Followed by a first person pronoun

    np_acts-logits-general · gemini-2.0-flash

    personal thoughts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    setCellValue
    -0.53
    complexType
    -0.50
    Kaieteur
    -0.47
    XmlAccessorType
    -0.47
    jspx
    -0.47
     '{@
    -0.45
     beginnetje
    -0.45
    PEOPLE
    -0.44
    andExpect
    -0.41
    gettyimages
    -0.40
    POSITIVE LOGITS
     my
    1.05
     mine
    0.85
    僕は
    0.77
     ours
    0.76
     nekem
    0.76
    PerformLayout
    0.75
    my
    0.74
     akong
    0.74
     mijn
    0.73
    うちの
    0.72
    Activations Density 0.302%

    No Known Activations