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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 9-GEMMASCOPE-RES-16K
    6. 3476
    Prev
    Next
    INDEX
    Explanations

    instances of screaming or loud vocal expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    WireFormatLite
    -0.45
    ArrayAdapter
    -0.36
    irited
    -0.34
    GOTREF
    -0.34
    ymce
    -0.34
    fjspx
    -0.34
     touched
    -0.33
    RegressionTest
    -0.32
     Patt
    -0.32
     Nop
    -0.32
    POSITIVE LOGITS
     scream
    1.53
     yell
    1.52
     shout
    1.48
     shouting
    1.43
     shouted
    1.43
     screaming
    1.42
     yelling
    1.41
     yelled
    1.40
     screamed
    1.39
     yells
    1.38
    Activations Density 0.283%

    No Known Activations