© 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
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 2265
    Prev
    Next
    INDEX
    Explanations

    descriptions of positive personal interactions and friendly character traits

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

    positive descriptive adjectives about people, like "cool", "nice", "friendly", "great".

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     planification
    -0.36
     minist
    -0.35
    PostInfinity
    -0.34
     Dhabi
    -0.33
    Catalog
    -0.33
     mas
    -0.33
    不了
    -0.32
    ietf
    -0.32
     grooming
    -0.32
     Jubilee
    -0.32
    POSITIVE LOGITS
    friendly
    0.73
     friendly
    0.72
     ramah
    0.64
    Friendly
    0.63
     vriende
    0.62
     Friendly
    0.60
    enderror
    0.60
     estekak
    0.59
     cordial
    0.59
    posedge
    0.57
    Activations Density 0.100%

    No Known Activations