© 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-2-9B-IT
    4. Residual Stream - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 114594
    Prev
    Next
    INDEX
    Explanations

    functions or methods that perform actions or operations

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

    to + verb

    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
     الحره
    -0.80
     estekak
    -0.68
    posedge
    -0.66
     tartalomajánló
    -0.65
     Réponses
    -0.65
    Παραπομπές
    -0.63
     kasarigan
    -0.62
    \{\\
    -0.60
    lgari
    -0.59
     itſelf
    -0.56
    POSITIVE LOGITS
     margin
    0.40
     rest
    0.38
     signed
    0.35
     hooked
    0.35
    二是
    0.34
     red
    0.33
     micro
    0.32
     two
    0.32
     non
    0.32
     pack
    0.31
    Activations Density 0.042%

    No Known Activations