Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 20-GEMMASCOPE-RES-16K
    6. 12591
    Prev
    Next
    INDEX
    Explanations

    the presence of specific formatting or beginning markers in a text, indicating a structural or programming context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    ↵↵
    -0.66
    ​
    -0.62
    <eos>
    -0.58
    ...
    -0.54
     while
    -0.54
     many
    -0.49
    z
    -0.49
     and
    -0.49
    …
    -0.48
    ’
    -0.48
    POSITIVE LOGITS
    __':
    
    0.98
    ImageContext
    0.92
     للاسماء
    0.90
     متعلقه
    0.88
    évaluateur
    0.86
    ErrUnexpectedEOF
    0.84
    TagMode
    0.84
    出版年
    0.83
    __":
    
    0.83
    ]")]
    0.83
    Activations Density 0.003%

    No Known Activations