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. 18-GEMMASCOPE-RES-16K
    6. 3877
    Prev
    Next
    INDEX
    Explanations

    narrative elements and perspectives from a first-person point of view in storytelling

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

    narration or conversation

    np_acts-logits-general · gemini-2.0-flash

    narration and dialogue

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    djangoproject
    -0.49
     caoutchouc
    -0.42
     Inscrivez
    -0.41
     reik
    -0.41
     termine
    -0.40
    ọn
    -0.40
     sinal
    -0.40
     amminist
    -0.40
    兆
    -0.39
    ssus
    -0.38
    POSITIVE LOGITS
     narrator
    0.96
    parsedMessage
    0.91
    narr
    0.91
     narr
    0.90
     dialogue
    0.89
     narration
    0.87
     dialogues
    0.83
     voice
    0.82
     Dialogue
    0.81
     interact
    0.81
    Activations Density 0.324%

    No Known Activations