© 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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 105760
    Prev
    Next
    INDEX
    Explanations

    terms related to various forms of visual representation and analysis in the arts and sciences, particularly focusing on photography and microscopy

    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
    (
    -0.68
    <
    -0.52
    Find
    -0.47
    ((
    -0.44
    Local
    -0.42
    (<
    -0.41
    Set
    -0.41
    Un
    -0.40
    ==
    -0.40
    ,(
    -0.40
    POSITIVE LOGITS
    iology
    0.91
    logy
    0.87
    ology
    0.85
    roscopy
    0.85
    ecology
    0.82
    mology
    0.82
    ography
    0.80
    graphy
    0.79
    pology
    0.79
    lography
    0.75
    Activations Density 0.036%

    No Known Activations