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. 4-GEMMASCOPE-RES-16K
    6. 10955
    Prev
    Next
    INDEX
    Explanations

    phrases related to permanence and long-lasting effects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Geplaatst
    -0.88
     HasFactory
    -0.77
    ImageContext
    -0.71
    RTLR
    -0.67
     richt
    -0.66
    """
    
    -0.64
    äu
    -0.61
    ":[{
    -0.61
     isoli
    -0.60
    PyTuple
    -0.60
    POSITIVE LOGITS
     permanent
    2.88
     Permanent
    2.60
    permanent
    2.57
     permanently
    2.55
    Permanent
    2.49
     PERMAN
    2.26
    perman
    2.21
     Perman
    2.20
     permanente
    2.15
     perman
    2.06
    Activations Density 0.104%

    No Known Activations