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. 2-GEMMASCOPE-RES-16K
    6. 15137
    Prev
    Next
    INDEX
    Explanations

    the presence of conjunctions or prepositions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Cane
    -0.57
     naturelles
    -0.56
     basse
    -0.55
     primario
    -0.55
    της
    -0.53
    jména
    -0.52
    selves
    -0.52
    Важно
    -0.52
     cuánto
    -0.52
     társ
    -0.52
    POSITIVE LOGITS
    endphp
    0.86
     محفوظة
    0.84
    encodeWith
    0.82
    IBarButtonItem
    0.80
    readFileSync
    0.80
    SBATCH
    0.75
    ˏ
    0.74
    initWith
    0.74
    HasAnnotation
    0.73
    Obrázky
    0.73
    Activations Density 0.049%

    No Known Activations