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. 25-GEMMASCOPE-RES-16K
    6. 3236
    Prev
    Next
    INDEX
    Explanations

    terms related to data analysis and interpretations of results

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    expandindo
    -0.60
    royd
    -0.59
    haustible
    -0.56
     Normdatei
    -0.53
     contextLoads
    -0.52
    SourceChecksum
    -0.49
    躇
    -0.49
     engraçadas
    -0.49
    InjectAttribute
    -0.48
    väg
    -0.47
    POSITIVE LOGITS
     match
    1.71
     matches
    1.68
     matched
    1.45
    match
    1.42
     matching
    1.41
     coincide
    1.38
     MATCH
    1.35
     Matches
    1.35
     Match
    1.35
    matches
    1.35
    Activations Density 1.076%

    No Known Activations