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. 11-GEMMASCOPE-RES-16K
    6. 7327
    Prev
    Next
    INDEX
    Explanations

    questions and statements concerning verification or validation of information

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Hook Layer
    11
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     مشين
    -0.68
     shalt
    -0.61
    most
    -0.61
    EndContext
    -0.58
     Daven
    -0.56
    Geplaatst
    -0.56
     pleaſure
    -0.55
     espa
    -0.54
     createState
    -0.54
    ContentAsync
    -0.53
    POSITIVE LOGITS
    それとも
    0.89
     whether
    0.89
    有没有
    0.81
     or
    0.80
     WHETHER
    0.80
    有沒有
    0.76
    是否有
    0.75
    是否
    0.74
     apakah
    0.74
    会不会
    0.73
    Activations Density 0.497%

    No Known Activations