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. 20-GEMMASCOPE-RES-16K
    6. 468
    Prev
    Next
    INDEX
    Explanations

    terms related to scientific and technical processes

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

    followed by a closing parenthesis

    np_acts-logits-general · gemini-2.0-flash

    words followed by punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    '}>
    -1.35
    ."));
    -1.33
    。】
    -1.26
    });*/
    -1.25
    .");
    
    -1.23
    .}}
    -1.21
    .";
    
    -1.21
    "}>
    -1.18
    }*/
    
    -1.16
    */
    
    
    -1.16
    POSITIVE LOGITS
    )
    1.33
    ),
    0.96
    ”
    0.88
    </code>
    0.78
    </
    0.77
    "
    0.74
    </strong>
    0.72
    ",
    0.72
    ”,
    0.66
    }
    0.62
    Activations Density 1.354%

    No Known Activations