© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 84679
    Prev
    Next
    INDEX
    Explanations

    keywords related to filesystem states and error handling

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

    isNotEmpty or empty

    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
    __":
    
    -0.51
    __":
    -0.47
    __':
    
    -0.46
    })();
    
    -0.43
    __':
    -0.42
    })$}
    -0.42
     *
    
    
    -0.42
    ExtendWith
    -0.42
    <?
    -0.41
    Xna
    -0.40
    POSITIVE LOGITS
    NotBlank
    1.70
    NotEmpty
    1.36
    notEmpty
    1.02
    isNotBlank
    0.80
    isNotEmpty
    0.77
    IsNotEmpty
    0.76
    OrWhiteSpace
    0.71
    oa̍t
    0.56
    aarrggbb
    0.52
    GEBURTSDATUM
    0.50
    Activations Density 0.001%

    No Known Activations