© 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 - 16k
    9. 38-GEMMASCOPE-RES-16K
    11. 10007
    Prev
    Next
    INDEX
    Explanations

    occurrences of personal pronouns and variations of the verb "to be."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     lijn
    -0.59
     bezoekers
    -0.58
     buurt
    -0.57
     tijden
    -0.55
     znajdują
    -0.54
     aż
    -0.54
     toekomst
    -0.54
     strijd
    -0.54
     kvinder
    -0.53
     zugesch
    -0.52
    POSITIVE LOGITS
     pu
    0.42
     '\\;'
    0.40
     mis
    0.40
     eu
    0.38
    featureID
    0.38
     domesti
    0.36
    exitRule
    0.34
     distribu
    0.32
    consulté
    0.32
     calcul
    0.32
    Activations Density 0.104%

    No Known Activations