© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 37-GEMMASCOPE-RES-16K
    6. 4007
    Prev
    Next
    INDEX
    Explanations

    phrases related to searching for or discovering something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.37.hook_resid_post
    Hook Layer
    37
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Expectations
    -0.44
     expectation
    -0.44
     Expectations
    -0.44
     expectancy
    -0.43
     erwarten
    -0.43
     expectations
    -0.42
    EXPECT
    -0.41
     constater
    -0.41
     EXPECT
    -0.41
     BoxFit
    -0.41
    POSITIVE LOGITS
     suitable
    1.11
     geeigne
    0.96
    suitable
    0.96
     подходя
    0.91
    合适的
    0.84
     passenden
    0.84
     answers
    0.84
    Suitable
    0.84
     Suitable
    0.83
     geschikt
    0.81
    Activations Density 0.539%

    No Known Activations