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. 15-GEMMASCOPE-RES-16K
    6. 4977
    Prev
    Next
    INDEX
    Explanations

    terms related to engines and their specifications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     nicio
    -0.82
     directos
    -0.78
     Teri
    -0.76
     Waw
    -0.74
     Darryl
    -0.72
     fêtes
    -0.72
     Rosewood
    -0.71
     Quant
    -0.70
     Draht
    -0.69
    +#+#
    -0.69
    POSITIVE LOGITS
     engines
    1.28
     Engines
    1.10
    ENGINE
    1.07
     engine
    1.05
    engines
    1.05
    Engine
    0.99
     ENGINE
    0.99
    engine
    0.95
    Engines
    0.95
     Engine
    0.95
    Activations Density 0.079%

    No Known Activations