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. 12813
    Prev
    Next
    INDEX
    Explanations

    legal terms and phrases related to statutes, regulations, and legal requirements

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

    Legal codes, statutes, and sections

    np_acts-logits-general · gemini-2.0-flash

    legal requirements and provisions

    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
     BorderRadius
    -0.65
    TypedDataSet
    -0.61
    enumi
    -0.60
    UnknownFields
    -0.56
    WindowConstants
    -0.52
    ReusableCell
    -0.50
     remarque
    -0.50
     Conducted
    -0.48
     exhibiting
    -0.48
     understatement
    -0.47
    POSITIVE LOGITS
     prohibit
    0.84
     authorizing
    0.84
     authorizes
    0.83
     authorize
    0.82
     stip
    0.80
     forbid
    0.79
     stipul
    0.78
     specify
    0.78
     prohibits
    0.77
     specifies
    0.77
    Activations Density 0.808%

    No Known Activations