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. 22-GEMMASCOPE-RES-16K
    6. 11348
    Prev
    Next
    INDEX
    Explanations

    elements related to psychological tension and character complexity

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

    Adverbs describing skill or quality

    np_acts-logits-general · gemini-2.0-flash

    skillful direction and performance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Popularity
    -0.50
     milhares
    -0.48
     théorie
    -0.48
     thousands
    -0.47
     famoso
    -0.47
     millions
    -0.47
     miglia
    -0.47
     obr
    -0.46
    stdc
    -0.46
     Actualmente
    -0.45
    POSITIVE LOGITS
     deft
    0.84
     admirably
    0.69
     smartly
    0.69
     ably
    0.69
     expertly
    0.67
     convincingly
    0.64
     Normdatei
    0.62
     skillfully
    0.60
    DockStyle
    0.60
     manages
    0.60
    Activations Density 0.315%

    No Known Activations