© 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-2-2B
    4. Residual Stream - 16k
    5. 17-GEMMASCOPE-RES-16K
    6. 8921
    Prev
    Next
    INDEX
    Explanations

    instances of the word "break" and its variations that indicate disruption or change

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.17.hook_resid_post
    Hook Layer
    17
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ơn
    -0.51
    BASEPATH
    -0.47
    ❋
    -0.46
     mouseY
    -0.45
     Glory
    -0.45
    帖最后由
    -0.45
    înd
    -0.44
    UTH
    -0.43
    Ceci
    -0.43
     zeros
    -0.42
    POSITIVE LOGITS
     break
    1.42
     breaking
    1.39
    break
    1.39
     BREAK
    1.34
     Break
    1.34
    Break
    1.32
     breaks
    1.32
     Breaks
    1.31
    breaking
    1.29
     Breaking
    1.28
    Activations Density 0.150%

    No Known Activations