© 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-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 90796
    Prev
    Next
    INDEX
    Explanations

    key-value pairs and data structures relevant to programming, particularly in API contexts

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

    javascript object properties

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     يتيمه
    -0.56
    OGND
    -0.47
     שוליים
    -0.46
     Occupied
    -0.44
    Occup
    -0.41
    BagConstraints
    -0.41
     prawda
    -0.41
     occupied
    -0.40
    المناصب
    -0.40
    AndEndTag
    -0.40
    POSITIVE LOGITS
    AddTagHelper
    0.50
    引
    0.43
     INTERESAR
    0.41
     Versch
    0.38
     Niedersch
    0.35
    Miejs
    0.35
    detectChanges
    0.35
     引
    0.35
    writeFieldEnd
    0.35
    اج
    0.35
    Activations Density 0.047%

    No Known Activations