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. 20-GEMMASCOPE-RES-16K
    6. 7692
    Prev
    Next
    INDEX
    Explanations

    elements of programming class structures and data definitions in code

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

    JSON data or responses

    np_acts-logits-general · gemini-2.0-flash

    json data structures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Hochspringen
    -0.56
    Clik
    -0.52
     userRepository
    -0.41
    շ
    -0.40
     capitales
    -0.40
     voyez
    -0.39
    userRepository
    -0.39
    UserRepository
    -0.39
     aldrig
    -0.38
     immaterial
    -0.37
    POSITIVE LOGITS
     snippetHide
    0.82
     data
    0.81
     success
    0.77
     result
    0.74
     message
    0.71
    Cyfeiriadau
    0.70
     doInBackground
    0.69
     SUCCESS
    0.69
     errno
    0.69
    success
    0.69
    Activations Density 0.140%

    No Known Activations