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. 22-GEMMASCOPE-RES-16K
    6. 8022
    Prev
    Next
    INDEX
    Explanations

    expressions related to personal growth and overcoming challenges

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

    Follows "have/has/had" and "used to"

    np_acts-logits-general · gemini-2.0-flash

    grown used to

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ComponentModel
    -0.59
     مواليد
    -0.53
     gouver
    -0.52
     solare
    -0.50
    ẹn
    -0.50
     지원
    -0.50
    蜒
    -0.49
    temon
    -0.49
     néz
    -0.48
    setValues
    -0.48
    POSITIVE LOGITS
     accustomed
    1.75
     habitu
    1.60
     acost
    1.55
     привы
    1.51
     habit
    1.46
     acostumb
    1.42
     familiar
    1.37
    慣
    1.35
     familiarity
    1.30
    习惯
    1.26
    Activations Density 0.292%

    No Known Activations