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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 20-GEMMASCOPE-RES-16K
    6. 5269
    Prev
    Next
    INDEX
    Explanations

    variations of the word "blow" in different contexts, particularly relating to events of destruction or failure

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

    Variations of "blow" or "swollen"

    np_acts-logits-general · gemini-2.0-flash

    blow, blowing, blown

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Normdatei
    -0.85
    QMetaType
    -0.67
    djangoproject
    -0.63
    WireFormatLite
    -0.61
    出版年
    -0.58
    sidemargin
    -0.58
    hicles
    -0.57
     IndexPath
    -0.57
    GOTREF
    -0.56
     JADX
    -0.56
    POSITIVE LOGITS
     blown
    1.20
     blew
    1.13
    blown
    1.03
     blow
    1.02
    Blow
    1.02
     Blow
    1.01
     blowing
    0.99
     blows
    0.99
    blowing
    0.86
    吹
    0.83
    Activations Density 0.067%

    No Known Activations