© 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. 14054
    Prev
    Next
    INDEX
    Explanations

    mentions of training, teamwork, and preparation contexts in athletic scenarios

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

    "training" or working out

    np_acts-logits-general · gemini-2.0-flash

    training and practice

    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
    JsonProperty
    -0.47
    tomatik
    -0.47
    gawai
    -0.46
     ramah
    -0.46
    ipolar
    -0.43
     acabados
    -0.42
     momentarily
    -0.42
     default
    -0.41
    낸
    -0.41
    ptid
    -0.41
    POSITIVE LOGITS
     training
    1.28
    training
    1.12
    Training
    1.07
     workouts
    1.06
     Training
    1.05
     practicing
    1.03
     trainings
    0.99
     practice
    0.98
     practising
    0.97
     TRAINING
    0.96
    Activations Density 0.158%

    No Known Activations