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. 82815
    Prev
    Next
    INDEX
    Explanations

    mentions of specific points in a given context or discussion

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

    point on, of, where, or S

    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
     pelúcia
    -0.57
     NavController
    -0.52
     dziecię
    -0.50
    })*/
    -0.50
     maioria
    -0.49
     doros
    -0.49
    surgical
    -0.48
     nadat
    -0.48
     sereia
    -0.48
     uitges
    -0.48
    POSITIVE LOGITS
     point
    1.83
    point
    1.75
     Point
    1.67
    Point
    1.66
     POINT
    1.59
    POINT
    1.51
     points
    1.24
    points
    1.17
     punto
    1.17
     Points
    1.12
    Activations Density 0.039%

    No Known Activations