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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 27-GEMMASCOPE-RES-16K
    6. 10466
    Prev
    Next
    INDEX
    Explanations

    data related to years and temporal references

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.27.hook_resid_post
    Hook Layer
    27
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    CWE
    -0.40
     abzu
    -0.35
    cerna
    -0.34
    kaŭ
    -0.33
     Loew
    -0.32
     Madsen
    -0.31
    envolvimento
    -0.30
    ripsi
    -0.29
    agner
    -0.29
    Джерела
    -0.29
    POSITIVE LOGITS
    Year
    1.09
     Year
    1.08
     year
    0.96
     YEAR
    0.96
    YEAR
    0.95
    year
    0.90
    getYear
    0.86
     otomatig
    0.81
     Years
    0.78
     année
    0.76
    Activations Density 0.140%

    No Known Activations