© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 24-GEMMASCOPE-RES-16K
    11. 12602
    Prev
    Next
    INDEX
    Explanations

    information related to model loading and error handling in programming

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

    Code, training, and model saving/loading

    np_acts-logits-general · gemini-2.0-flash

    training models and loss

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Refuse
    -0.46
    orem
    -0.46
     couvercle
    -0.46
     nonatomic
    -0.45
    xase
    -0.45
     refuse
    -0.43
    ăzi
    -0.42
     room
    -0.42
    Javadoc
    -0.41
    嫁
    -0.41
    POSITIVE LOGITS
     train
    0.84
     trained
    0.83
     training
    0.82
     loss
    0.81
     trainer
    0.77
    train
    0.77
    Train
    0.76
     Train
    0.76
     model
    0.75
     Loss
    0.74
    Activations Density 0.335%

    No Known Activations