© 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-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 14135
    Prev
    Next
    INDEX
    Explanations

    highly positive evaluations of food textures and flavors

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

    Describes textures or qualities

    np_acts-logits-general · gemini-2.0-flash

    soft, tender, flaky, crispy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Autoritní
    -0.67
     nakalista
    -0.58
    openzeppelin
    -0.54
     zebrafish
    -0.50
     weeding
    -0.48
     בנושא
    -0.48
    -------
    -0.48
     Claro
    -0.48
     diversi
    -0.48
     Chriftian
    -0.47
    POSITIVE LOGITS
     soft
    1.26
     softness
    1.21
     Soft
    1.11
    soft
    1.10
     softer
    1.07
    Soft
    1.05
     SOFT
    1.05
    SOFT
    1.00
     tender
    0.93
     lembut
    0.89
    Activations Density 0.101%

    No Known Activations