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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 57319
    Prev
    Next
    INDEX
    Explanations

    scientific terminology related to measurement and evaluation in research contexts

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

    Technical terms related to data processing, analysis, or filtering in scientific and research contexts, often appearing with punctuation marks such as commas or periods nearby.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    providedIn
    -0.49
    RTEX
    -0.48
     ivelany
    -0.47
    BeginInit
    -0.44
     ErrInvalid
    -0.44
     للاسماء
    -0.43
    EndInit
    -0.43
     newOwner
    -0.43
    endphp
    -0.43
    Hypothesis
    -0.42
    POSITIVE LOGITS
    RenderAtEndOf
    0.44
    ChildScrollView
    0.43
    MessageTagHelper
    0.43
    łgor
    0.41
     InputDecoration
    0.38
     mxArray
    0.36
     gerne
    0.36
    MarshalTo
    0.36
    直径
    0.35
    Hentet
    0.34
    Activations Density 0.138%

    No Known Activations