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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 9863
    Prev
    Next
    INDEX
    Explanations

    references to the use of technology and devices in everyday life

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

    Frequency, often, or repeated actions

    np_acts-logits-general · gemini-2.0-flash

    frequently or often

    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
    первых
    -0.52
    marginVertical
    -0.49
    jsdelivr
    -0.48
    начала
    -0.48
     =>
    
    -0.46
    Erreferentziak
    -0.45
    NotFoundError
    -0.43
    året
    -0.43
    原始内容存档于
    -0.43
    ERG
    -0.42
    POSITIVE LOGITS
     frequently
    1.90
     alot
    1.63
     often
    1.61
    frequently
    1.61
     regularly
    1.56
     frequent
    1.55
     lots
    1.44
    frequent
    1.42
     Frequently
    1.40
     frequentemente
    1.38
    Activations Density 0.535%

    No Known Activations