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. 8530
    Prev
    Next
    INDEX
    Explanations

    mentions of the term "Hon" and its variations, reflecting a focus on references to honorific titles or names

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

    "Hon" or "Honor" (or variants) followed by other words

    np_acts-logits-general · gemini-2.0-flash

    honorable

    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
     estekak
    -0.71
    Sucesor
    -0.71
    Искәрмәләр
    -0.67
     utafitiHapana
    -0.61
     ThemeData
    -0.56
    Ligações
    -0.53
     inne
    -0.52
     pinulongan
    -0.52
     spares
    -0.51
    izmu
    -0.50
    POSITIVE LOGITS
     HON
    1.08
    HON
    1.03
     HONOR
    0.96
    hon
    0.92
     dishon
    0.87
    Hon
    0.86
    Honor
    0.84
     Hon
    0.84
     hon
    0.84
     onor
    0.82
    Activations Density 0.061%

    No Known Activations