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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 12106
    Prev
    Next
    INDEX
    Explanations

    phrases indicating agreement or disagreement in research findings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     المعيارى
    -0.70
     GenerationType
    -0.60
    featureID
    -0.59
    Дерекк
    -0.56
    HasForeignKey
    -0.56
    jsxFileName
    -0.55
    évaluateur
    -0.54
    AddTagHelper
    -0.54
     AssemblyCulture
    -0.54
     autorytatywna
    -0.52
    POSITIVE LOGITS
    awtextra
    0.40
    Obviously
    0.38
    RSpec
    0.36
     imidlertid
    0.34
     relatively
    0.33
    毋
    0.32
    首先
    0.32
     Obviously
    0.32
    Unlike
    0.32
    Initially
    0.31
    Activations Density 3.342%

    No Known Activations