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. 1-GEMMASCOPE-RES-16K
    6. 14789
    Prev
    Next
    INDEX
    Explanations

    references to different times of day, particularly focused on the evening

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.1.hook_resid_post
    Hook Layer
    1
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hoeddwyd
    -0.66
    featureID
    -0.65
    ExecuteReader
    -0.64
     Exactos
    -0.64
    GIVEREF
    -0.61
    tonode
    -0.60
    ymce
    -0.60
     Grade
    -0.59
    oa̍t
    -0.58
    UObject
    -0.56
    POSITIVE LOGITS
     evening
    2.68
     afternoon
    2.48
    evening
    2.41
     morning
    2.40
     Evening
    2.28
    Evening
    2.20
    morning
    2.14
     EVENING
    2.07
     Afternoon
    2.06
    Afternoon
    2.04
    Activations Density 0.066%

    No Known Activations