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

    Neuronpedia

    APICircuit 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. 16-GEMMASCOPE-RES-16K
    6. 3086
    Prev
    Next
    INDEX
    Explanations

    terms related to directionality and orientation in various contexts

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

    "Direction" or "directed"

    np_acts-logits-general · gemini-2.0-flash

    direction, orientation, or flow

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    belec
    -0.76
     Jank
    -0.74
    ReusableCell
    -0.73
    ikkert
    -0.72
    ecore
    -0.69
    :✨
    -0.69
    fgaben
    -0.67
    mybatisplus
    -0.65
     Walton
    -0.65
    XtraGrid
    -0.65
    POSITIVE LOGITS
     directions
    1.06
     Direction
    0.96
     direction
    0.94
     DIRECTION
    0.94
     Directions
    0.92
     Toward
    0.88
    directions
    0.88
     directing
    0.87
    DIRECTION
    0.84
     Dir
    0.81
    Activations Density 0.207%

    No Known Activations