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. 12-GEMMASCOPE-RES-16K
    6. 12094
    Prev
    Next
    INDEX
    Explanations

    scientific and technical terms related to experimental results and methodologies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     however
    -0.84
    however
    -0.75
     and
    -0.72
     However
    -0.72
    However
    -0.71
     porém
    -0.69
     però
    -0.62
     aber
    -0.61
     jedoch
    -0.61
     însă
    -0.58
    POSITIVE LOGITS
     dermed
    0.76
    OGND
    0.64
     consequently
    0.62
     виправивши
    0.62
     therewith
    0.61
    Enllaces
    0.60
     thereof
    0.60
    انجليز
    0.60
     accordingly
    0.59
     därför
    0.59
    Activations Density 1.854%

    No Known Activations