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. 18-GEMMASCOPE-RES-16K
    6. 5254
    Prev
    Next
    INDEX
    Explanations

    words related to seeking attention or support

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

    after "the" or possessive "'s"

    np_acts-logits-general · gemini-2.0-flash

    earning confidence

    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
    @[+][
    -0.65
    IBOutlet
    -0.65
    enderror
    -0.61
    сылкі
    -0.59
    JsonValue
    -0.58
    matchCondition
    -0.58
    ANTLR
    -0.57
    setCharacter
    -0.57
    الدراسه
    -0.56
    IBarButtonItem
    -0.55
    POSITIVE LOGITS
     attention
    1.43
    attention
    1.18
     Attention
    1.15
    Attention
    1.07
     ATTENTION
    1.01
     attentions
    0.95
     atención
    0.93
     atenção
    0.87
     Aufmerksamkeit
    0.86
     aandacht
    0.85
    Activations Density 0.172%

    No Known Activations