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
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 53918
    Prev
    Next
    INDEX
    Explanations

    attributes related to friendliness and approachability

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    -0.88
    DockStyle
    -0.74
     GenerationType
    -0.66
    principalColumn
    -0.66
     JAXBElement
    -0.66
    ंदीखरीदारी
    -0.65
    yntaxException
    -0.65
    таратура
    -0.63
     tartalomajánló
    -0.62
     kasarigan
    -0.61
    POSITIVE LOGITS
     friendly
    0.52
     smiling
    0.48
     souri
    0.44
     freundlichen
    0.42
     smile
    0.42
     Smiling
    0.41
     warm
    0.41
    Smiling
    0.41
     warmly
    0.40
     smiles
    0.39
    Activations Density 0.005%

    No Known Activations