© 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. 3-GEMMASCOPE-RES-16K
    6. 1966
    Prev
    Next
    INDEX
    Explanations

    operations related to stack manipulation and queue processing

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    就跟
    -0.46
    withIdentifier
    -0.43
    __(
    -0.41
     Dun
    -0.40
    asList
    -0.39
     identi
    -0.39
    bald
    -0.39
    のが
    -0.38
     Main
    -0.38
     Mr
    -0.38
    POSITIVE LOGITS
    PerformLayout
    1.04
    ImageContext
    0.92
     الاطلاع
    0.88
    WaitGroup
    0.85
     dequeue
    0.84
    dequeue
    0.82
    featureID
    0.82
     Exacts
    0.80
    pleft
    0.79
    )++;
    0.79
    Activations Density 0.459%

    No Known Activations