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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 19-GEMMASCOPE-TRANSCODER-16K
    4. 10165
    Prev
    Next
    INDEX
    Explanations

    World of Warcraft-related terms regarding character stats and abilities.

    oai_token-act-pair · gemini-2.0-flash

    Various linguistic patterns observed including: idioms, comparative word endings, possessive nouns, programming code tokens, grammatical markers, and computational/mathematical terminology.

    eleuther_acts_top20 · claude-3-5-haiku-20241022Triggered by @oreo4455

    numerical values, especially years or dates, particularly in formal or technical documents.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @oreo4455

    Programming language syntax elements and control structures in code snippets, including function parameters, loop constructs, and variable references.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @oreo4455

    short words or word segments, particularly those with numeric or symbolic characters.

    oai_token-act-pair · deepseek-v3Triggered by @oreo4455
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope/layer_19/width_16k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     kaarangay
    -0.62
     defe
    -0.59
    ConstraintMaker
    -0.54
    ]<<"
    -0.50
     createSlice
    -0.49
     Taschen
    -0.48
     forState
    -0.48
    Composable
    -0.47
    InitStruct
    -0.46
    ResumeLayout
    -0.46
    POSITIVE LOGITS
    MLLoader
    0.66
    matician
    0.54
    InjectAttribute
    0.53
    xase
    0.53
    Pilot
    0.53
     lèvres
    0.52
     BrowserModule
    0.51
    orestation
    0.51
    RTEX
    0.50
     Drapeau
    0.50
    Activations Density 0.316%

    No Known Activations