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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 122774
    Prev
    Next
    INDEX
    Explanations

    mentions of living rooms or parlors in houses or apartments.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @subhashk

    references to specific rooms in a house, particularly living rooms and parlors.

    oai_token-act-pair · gpt-4oTriggered by @joshengels

    mentions of living rooms or similar domestic spaces

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    XmlIgnore
    -0.54
    Personensuche
    -0.48
     Brazil
    -0.43
     bré
    -0.39
    Brazil
    -0.38
    Australie
    -0.38
    assertThrows
    -0.37
    Clave
    -0.36
     Korea
    -0.36
     Pedro
    -0.36
    POSITIVE LOGITS
     Wohnzimmer
    0.74
     InputDecoration
    0.71
     lounge
    0.71
    lounge
    0.68
     sofa
    0.67
    Sofa
    0.66
    客厅
    0.66
     BoxDecoration
    0.65
     couch
    0.65
     Lounge
    0.64
    Activations Density 0.005%

    No Known Activations