© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 3931
    Prev
    Next
    INDEX
    Explanations

    information related to software licenses and assembly metadata

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/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.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Sucesor
    -0.59
    وعة
    -0.53
    Према
    -0.49
     allá
    -0.48
    发表于
    -0.47
    CAPTURE
    -0.47
    dismo
    -0.47
     flamingo
    -0.45
    ⓧ
    -0.45
    ösung
    -0.45
    POSITIVE LOGITS
    InteropServices
    0.74
    ]='\
    0.57
    bibinfo
    0.56
    ).__
    0.56
    autique
    0.55
    bibfield
    0.55
    CompilerServices
    0.54
    íslu
    0.54
    исленность
    0.54
     ioutil
    0.53
    Activations Density 0.018%

    No Known Activations