© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 65k
    9. 20-GEMMASCOPE-RES-65K
    11. 677
    Prev
    Next
    INDEX
    Explanations

    the intensifier "really" and its variations in emotional or evaluative contexts

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

    intensifiers or adverbs that suggest emphasis or significance

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

    really followed by descriptor

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lamabad
    -0.65
    Chaos
    -0.58
    matchCondition
    -0.56
     depoz
    -0.55
    fois
    -0.54
     atheism
    -0.51
     hallucinations
    -0.51
    ctp
    -0.51
    
    -0.50
     Fais
    -0.50
    POSITIVE LOGITS
     nice
    0.67
    mergeFrom
    0.66
     cool
    0.65
    MessageOf
    0.64
     interesting
    0.62
    expandindo
    0.61
    MergeFrom
    0.61
     good
    0.60
     great
    0.59
     only
    0.59
    Activations Density 0.019%

    No Known Activations