© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 131k
    9. 24-GEMMASCOPE-RES-131K
    11. 627
    Prev
    Next
    INDEX
    Explanations

    quantitative measurements and estimates related to size, counts, and scale in various contexts

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

    quantifying amounts or scaled numbers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    úgó
    -0.44
    LabelTagHelper
    -0.40
     lenker
    -0.39
     Economy
    -0.38
     Simp
    -0.36
     Forced
    -0.35
     Sheep
    -0.35
     enc
    -0.34
    !*\
    -0.34
    unaga
    -0.34
    POSITIVE LOGITS
     dizaines
    0.73
     decenas
    0.70
     cientos
    0.69
     ratusan
    0.67
     hundreds
    0.67
    Hundreds
    0.67
     dozens
    0.66
    hundreds
    0.65
     ribuan
    0.64
     thousands
    0.62
    Activations Density 0.039%

    No Known Activations