Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 67539
    Prev
    Next
    INDEX
    Explanations

    legal terminology and references to court proceedings

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

    testimony and recounting events

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    IntoConstraints
    -0.73
    MemoryWarning
    -0.64
    ImageContext
    -0.57
    ataire
    -0.54
     myſelf
    -0.53
    MigrationBuilder
    -0.52
    SerializedName
    -0.52
     convertView
    -0.51
    stateProvider
    -0.49
    Portale
    -0.49
    POSITIVE LOGITS
    addCriterion
    0.42
     recount
    0.39
     recalled
    0.39
     recall
    0.38
     recollect
    0.37
     recounts
    0.36
    UnknownFieldSet
    0.36
    ">//
    0.35
     REC
    0.33
     retrospect
    0.33
    Activations Density 0.336%

    No Known Activations