© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 1670
    Prev
    Next
    INDEX
    Explanations

    complexities and challenges in diagnosis or distinction among similar entities

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

    complex phrases involving difficulty or ease of distinguishing between things.

    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
     kasarigan
    -0.55
    antwort
    -0.50
    AnchorStyles
    -0.50
    Consistent
    -0.48
     einverstanden
    -0.48
    rückt
    -0.46
     المعيارى
    -0.46
     ContentValues
    -0.45
    yayım
    -0.45
     CallOverrides
    -0.45
    POSITIVE LOGITS
     visually
    0.52
     distinguishable
    0.46
     indistingu
    0.43
     visuel
    0.42
     distinguishing
    0.41
     distinguer
    0.41
     confund
    0.40
     easily
    0.40
     Easily
    0.39
     confused
    0.38
    Activations Density 0.338%

    No Known Activations