© 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. 22-GEMMASCOPE-RES-16K
    6. 15479
    Prev
    Next
    INDEX
    Explanations

    present tense verbs indicating existence or state of being

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

    Following "are," "is," or "are always."

    np_acts-logits-general · gemini-2.0-flash

    allowed, required, or free

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     kasarigan
    -0.52
    ReusableCell
    -0.51
     proprement
    -0.51
    AnimationsModule
    -0.51
    مصادر
    -0.51
    ETCH
    -0.48
    Namara
    -0.46
    afficheront
    -0.46
    toMatchSnapshot
    -0.46
    jalna
    -0.46
    POSITIVE LOGITS
     free
    1.35
    free
    1.19
     welcome
    1.12
     entitled
    1.01
    Free
    0.98
     required
    0.97
     allowed
    0.95
    welcome
    0.94
     Free
    0.94
     libre
    0.89
    Activations Density 0.259%

    No Known Activations