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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 5446
    Prev
    Next
    INDEX
    Explanations

    the use of specific programming language syntax related to error handling and exceptions

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

    error handling code structures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     pleaſure
    -0.56
     Houſe
    -0.52
     ſta
    -0.50
     myſelf
    -0.49
     виправивши
    -0.49
     houſe
    -0.49
     Majefty
    -0.48
     '\\;'
    -0.47
    出版年
    -0.45
    ſelves
    -0.44
    POSITIVE LOGITS
     astore
    0.69
    Personendaten
    0.50
    clonal
    0.50
    脚注の使い方
    0.47
    Билгалдахарш
    0.47
     diagnose
    0.47
    BufferException
    0.46
     DIAG
    0.44
    Einzelnachweise
    0.44
     autorytatywna
    0.43
    Activations Density 0.001%

    No Known Activations