© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 9-GEMMASCOPE-RES-131K
    11. 84157
    Prev
    Next
    INDEX
    Explanations

    spaces or non-words used in mathematical equations or notation

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

    ```json[ {"text": "code blocks", "score": 0.9732549071311951}, {"text": "javascript code", "score": 0.9550331234931946}, {"text": "css rules", "score": 0.9340117574977875}, {"text": "html tags", "score": 0.9216808080673218}, {"text": "latex commands", "score": 0.8756493926048279}, {"text": "python code", "score": 0.8675746846199036}, {"text": "function definitions", "score": 0.8404641151428223}, {"text": "markup language", "score": 0.8317321586608887}, {"text": "program code", "score": 0.8094831037521362}, {"text": "code snippets", "score": 0.799100935459137}, {"text": "programming languages", "score": 0.7810649871826172}, {"text": "scripting languages", "score": 0.7787901163101196}, {"text": "code examples", "score": 0.7742959213256836}, {"text": "commands", "score": 0.7727951407432556}, {"text": "syntax", "score": 0.763621985912323}, {"text": "programming", "score": 0.7563790082645416}, {"text": "keywords", "score": 0.7464627027511597}, {"text": "regex", "score": 0.73064124584198}, {"text": "variables", "score": 0.729516279220581}, {"text": "declarations", "score": 0.7237746000289917}]```

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Referències
    -0.54
     يتيمه
    -0.53
    ConstraintMaker
    -0.50
     GoogleFonts
    -0.50
    pedidos
    -0.48
    uque
    -0.48
     InputDecoration
    -0.48
    extAlignment
    -0.47
     maritime
    -0.47
    bearded
    -0.46
    POSITIVE LOGITS
    PerformLayout
    0.47
    SharedCtor
    0.47
    </h6>
    0.44
    ]")]
    0.44
     ""],
    0.44
    </h5>
    0.43
    EndInit
    0.42
    </h4>
    0.41
     disambiguazione
    0.41
    ″]
    0.40
    Activations Density 0.009%

    No Known Activations