© 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. 9-GEMMASCOPE-RES-16K
    6. 209
    Prev
    Next
    INDEX
    Explanations

    negative descriptors or concepts related to "bad."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    AndEndTag
    -0.89
    WireFormatLite
    -0.88
     useRouter
    -0.87
     Mukherjee
    -0.86
     CRS
    -0.86
     copii
    -0.84
    adaptiveStyles
    -0.83
    Coppa
    -0.82
     tourmaline
    -0.82
    WithFormat
    -0.82
    POSITIVE LOGITS
     Bad
    1.57
     BAD
    1.55
    Bad
    1.51
     bad
    1.43
    bad
    1.41
    BAD
    1.39
     Baden
    0.80
     Bader
    0.76
     good
    0.74
    good
    0.74
    Activations Density 0.040%

    No Known Activations