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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 23-GEMMASCOPE-RES-16K
    6. 6435
    Prev
    Next
    INDEX
    Explanations

    subjects and adjectives related to positive and professional experiences in service contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.23.hook_resid_post
    Hook Layer
    23
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     paraf
    -0.52
     équilibr
    -0.49
     stabilisation
    -0.48
    jeras
    -0.46
     newOwner
    -0.46
     juridiques
    -0.45
     équilibre
    -0.45
     duele
    -0.45
     asymmetric
    -0.45
     stabilization
    -0.45
    POSITIVE LOGITS
     friendly
    1.01
     understanding
    0.95
     accommodating
    0.93
     courteous
    0.92
     helpful
    0.88
     polite
    0.86
     welcoming
    0.81
    understanding
    0.78
     knowledgeable
    0.78
     Friendly
    0.78
    Activations Density 0.087%

    No Known Activations