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-2-9B-IT
    4. Residual Stream - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 74963
    Prev
    Next
    INDEX
    Explanations

    positive descriptors and adjectives that convey a sense of quality or performance

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

    pretty adjective

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    XmlIgnore
    -0.42
    Personendaten
    -0.38
     CWE
    -0.38
    adaptiveStyles
    -0.38
     actionTypes
    -0.36
     invokingState
    -0.35
    cdk
    -0.35
    RotationOrder
    -0.35
    protoimpl
    -0.34
    actionMode
    -0.33
    POSITIVE LOGITS
     pretty
    1.09
    pretty
    1.08
     Pretty
    1.04
    Pretty
    1.04
     ziemlich
    0.80
     PRET
    0.73
     bastante
    0.70
     довольно
    0.68
    basically
    0.65
     Yeah
    0.64
    Activations Density 0.021%

    No Known Activations