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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 14591
    Prev
    Next
    INDEX
    Explanations

    terms related to suffering and health conditions

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

    pain, hardship, or negative experiences

    np_acts-logits-general · gemini-2.0-flash

    suffering and hardships

    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
    TypedDataSet
    -0.66
    ImageBackground
    -0.60
     jeito
    -0.57
     насељу
    -0.56
    esModule
    -0.55
    pyx
    -0.53
    crum
    -0.51
    Portale
    -0.51
    juje
    -0.50
     Pompey
    -0.50
    POSITIVE LOGITS
    AddTagHelper
    0.68
     difficulties
    0.67
    Personensuche
    0.63
     firsthand
    0.59
     considerable
    0.58
     backlash
    0.56
     đựng
    0.55
     tremendous
    0.55
     unknownFields
    0.55
     hardships
    0.55
    Activations Density 0.182%

    No Known Activations