Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 85553
    Prev
    Next
    INDEX
    Explanations

    various types of code syntax and structure elements

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

    structured data" delimited by |

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
     ſind
    -0.91
    niſſe
    -0.90
    ſehen
    -0.86
     queſta
    -0.82
    iſchen
    -0.80
    iſten
    -0.78
     iſt
    -0.78
    iſen
    -0.77
    majánló
    -0.77
     laſſen
    -0.75
    POSITIVE LOGITS
     וב
    0.53
     מל
    0.51
     כש
    0.49
    =
    0.48
     שב
    0.46
    -
    0.46
     وال
    0.46
     وك
    0.45
     כשה
    0.44
    __*/
    0.44
    Activations Density 0.352%

    No Known Activations