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. 24-GEMMASCOPE-RES-16K
    6. 13994
    Prev
    Next
    INDEX
    Explanations

    elements related to cultural complexities and narratives within Australian issues

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

    Exploration of themes/issues in literature

    np_acts-logits-general · gemini-2.0-flash

    thematic elements explored

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     propter
    -0.62
    iastical
    -0.58
     fubject
    -0.56
     purpoſe
    -0.56
     يتيمه
    -0.56
     Ahor
    -0.55
    zension
    -0.54
    unteer
    -0.54
     houſe
    -0.54
     Disadvantages
    -0.53
    POSITIVE LOGITS
     explored
    0.80
     interwoven
    0.76
     through
    0.73
     exploring
    0.71
     intertwined
    0.71
     while
    0.70
     fused
    0.69
     amidst
    0.69
     informed
    0.68
     melding
    0.68
    Activations Density 0.288%

    No Known Activations