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. 1525
    Prev
    Next
    INDEX
    Explanations

    references to herbs and related topics

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

    "Her" or words starting with "Her"

    np_acts-logits-general · gemini-2.0-flash

    words beginning with "Her"

    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
     vPvB
    -0.81
     Efq
    -0.80
     nahilalakip
    -0.80
     Lumpur
    -0.78
    Portale
    -0.74
     himſelf
    -0.73
     Normdatei
    -0.71
     Rys
    -0.71
    FieldBuilder
    -0.71
    enosis
    -0.70
    POSITIVE LOGITS
     Her
    1.22
     HER
    1.04
    Her
    1.04
     her
    0.96
    her
    0.90
    HER
    0.87
     Herzog
    0.68
     Heron
    0.64
     heron
    0.59
    Hers
    0.59
    Activations Density 0.086%

    No Known Activations