© 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
    4. Residual Stream - 16k
    5. 38-GEMMASCOPE-RES-16K
    6. 8894
    Prev
    Next
    INDEX
    Explanations

    assertions of truth

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MTG
    -0.45
    TagHelper
    -0.43
    mous
    -0.43
    omez
    -0.42
    stdc
    -0.41
     JpaRepository
    -0.41
    prefixer
    -0.41
     mouseY
    -0.41
    rict
    -0.41
    Bowling
    -0.41
    POSITIVE LOGITS
    False
    0.60
    FALSE
    0.54
     False
    0.54
     True
    0.52
    FirstResponder
    0.51
    True
    0.50
     believers
    0.50
    false
    0.50
     true
    0.47
     FALSE
    0.47
    Activations Density 0.150%

    No Known Activations