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-IT
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 15761
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to names and titles of individuals

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

    Numbers followed by punctuation

    np_acts-logits-general · gemini-2.0-flash

    aged resident

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Rè
    -0.44
     Meksiku
    -0.40
     Roskov
    -0.38
    Бахар
    -0.38
     kuta
    -0.38
    лтамалар
    -0.37
     LIABLE
    -0.37
    脚注の使い方
    -0.37
     suffices
    -0.35
     Allerdings
    -0.35
    POSITIVE LOGITS
     died
    0.80
     Passed
    0.70
     resident
    0.69
     beloved
    0.68
     peacefully
    0.65
     passed
    0.63
     tartalomajánló
    0.60
     formerly
    0.59
    Beloved
    0.59
    Died
    0.58
    Activations Density 0.021%

    No Known Activations