© 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. 3-GEMMASCOPE-RES-16K
    6. 8976
    Prev
    Next
    INDEX
    Explanations

    specific prefixes or suffixes in scientific and taxonomic terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.3.hook_resid_post
    Hook Layer
    3
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    waża
    -0.30
     ettei
    -0.28
    tutup
    -0.28
    chieht
    -0.28
    centerline
    -0.26
    roč
    -0.26
     Gegenteil
    -0.25
    zerű
    -0.24
    dijual
    -0.24
    espejo
    -0.24
    POSITIVE LOGITS
    ping
    1.17
    ped
    1.13
    pers
    1.12
    pd
    1.11
    pet
    1.10
    pac
    1.09
    py
    1.08
    pas
    1.06
    pan
    1.05
    pal
    1.05
    Activations Density 1.068%

    No Known Activations