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. 18-GEMMASCOPE-RES-16K
    6. 1062
    Prev
    Next
    INDEX
    Explanations

    relationships and conflicts within personal and social dynamics

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

    problems or solutions

    np_acts-logits-general · gemini-2.0-flash

    solve problems and troubles

    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
     WaitForSeconds
    -0.60
    IFICACIONES
    -0.54
    setAccessible
    -0.52
     antica
    -0.49
     kés
    -0.49
    DeleteBehavior
    -0.49
    fohlen
    -0.48
    absch
    -0.48
    --]
    -0.48
     marad
    -0.48
    POSITIVE LOGITS
     problems
    1.87
     troubles
    1.76
    problems
    1.70
     woes
    1.62
     Problems
    1.60
    Problems
    1.56
     PROBLEMS
    1.47
    PROBLE
    1.44
     problem
    1.43
    problem
    1.40
    Activations Density 0.226%

    No Known Activations