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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 14541
    Prev
    Next
    INDEX
    Explanations

    phrases indicating awareness or knowledge about specific subjects

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

    Follows verbs "be", "made", or similar

    np_acts-logits-general · gemini-2.0-flash

    state of awareness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 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
    entru
    -0.55
    Pty
    -0.54
    gink
    -0.54
     pory
    -0.54
    inaison
    -0.52
    timedelta
    -0.52
     Bergh
    -0.51
    Reprodução
    -0.51
    itage
    -0.49
    шеб
    -0.48
    POSITIVE LOGITS
     awareness
    2.52
     aware
    2.25
    Awareness
    2.15
     Awareness
    2.13
    awareness
    2.12
     Aware
    2.01
    aware
    1.84
     knowledge
    1.81
     consciousness
    1.79
     conscious
    1.65
    Activations Density 0.379%

    No Known Activations