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. 51254
    Prev
    Next
    INDEX
    Explanations

    references to SQL (Structured Query Language) and its variations

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

    SQL database language

    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
    <bos>
    -0.48
    Distribuzione
    -0.40
    <em>
    -0.39
     производстве
    -0.39
    penup
    -0.39
     BRO
    -0.35
    <strong>
    -0.35
     male
    -0.35
     featured
    -0.34
    chan
    -0.34
    POSITIVE LOGITS
    SQL
    2.16
     SQL
    2.00
     SQLITE
    1.01
     SQLAlchemy
    0.95
    Sql
    0.93
    sql
    0.93
     Sql
    0.93
     MySQL
    0.89
     sql
    0.84
     SqlCommand
    0.82
    Activations Density 0.004%

    No Known Activations