© 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. 16-GEMMASCOPE-RES-16K
    6. 1872
    Prev
    Next
    INDEX
    Explanations

    patterns related to error handling and control flow in coding

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

    Code snippets with "out" or "return" statements

    np_acts-logits-general · gemini-2.0-flash

    relationship between

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     sè
    -0.42
     Verhält
    -0.42
    زب
    -0.42
     marito
    -0.41
     Oppen
    -0.41
     surround
    -0.41
    unan
    -0.40
    тего
    -0.40
    mato
    -0.40
    stru
    -0.40
    POSITIVE LOGITS
    Hochspringen
    1.11
    AddTagHelper
    1.10
     مشين
    1.06
    rungsseite
    1.01
    findpost
    0.93
    writeFieldEnd
    0.92
    Билгалдахарш
    0.90
    脚注の使い方
    0.90
    Autoritní
    0.86
    RegressionTest
    0.85
    Activations Density 0.104%

    No Known Activations