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. 1404
    Prev
    Next
    INDEX
    Explanations

    code elements and constructs related to user interface (UI) components and properties

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

    `ft` or `tft` followed by other characters

    np_acts-logits-general · gemini-2.0-flash

    tft.add_text

    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
    hyrchwyd
    -0.57
     natomiast
    -0.48
     either
    -0.47
     inoltre
    -0.47
    @[+][
    -0.47
     någon
    -0.47
     아니라
    -0.46
     while
    -0.45
    IndentedString
    -0.45
     noget
    -0.43
    POSITIVE LOGITS
    mybatisplus
    1.03
    iprot
    0.89
    AndEndTag
    0.81
    TagMode
    0.75
    Билгалдахарш
    0.74
     שוליים
    0.72
     purpoſe
    0.71
    VolleyError
    0.70
     >=",
    0.69
    StandardCharsets
    0.69
    Activations Density 0.251%

    No Known Activations