© 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. 20-GEMMASCOPE-RES-16K
    6. 8330
    Prev
    Next
    INDEX
    Explanations

    phrases related to outcomes or conclusions

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

    "Turn" followed by "out"

    np_acts-logits-general · gemini-2.0-flash

    turn out

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     externi
    -0.53
    Calais
    -0.49
     CURIAM
    -0.49
    はじめに
    -0.49
     Unsc
    -0.49
     whet
    -0.47
    EXTERN
    -0.45
    UrlResolution
    -0.45
     Deposition
    -0.45
     clemen
    -0.45
    POSITIVE LOGITS
     out
    0.91
     BoxFit
    0.80
    SpringBootTest
    0.63
    DockStyle
    0.60
     مرئيه
    0.60
    StoryboardSegue
    0.59
    достатки
    0.59
    ArgsConstructor
    0.58
    ExtendWith
    0.57
    ViewSet
    0.57
    Activations Density 0.100%

    No Known Activations