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 - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 53520
    Prev
    Next
    INDEX
    Explanations

    instances of the verb "be" in various forms

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

    state of being

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     revol
    -0.46
    assertThrows
    -0.45
     sorrows
    -0.44
     VOS
    -0.44
     možno
    -0.44
     moustache
    -0.42
     Demos
    -0.42
     hydrological
    -0.41
     Rosh
    -0.41
    équi
    -0.41
    POSITIVE LOGITS
    存于互联网档案馆
    0.86
    tshell
    0.80
     utafitiHapana
    0.74
    MessageOf
    0.68
    ellate
    0.67
    eleste
    0.67
    GEBURTSDATUM
    0.67
     Савезне
    0.66
     صوتيه
    0.65
     समीक्षाओं
    0.65
    Activations Density 0.047%

    No Known Activations