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. 15399
    Prev
    Next
    INDEX
    Explanations

    emotional connections and sentiments in personal experiences

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

    "you," "us," or "me" and positive adverbs

    np_acts-logits-general · gemini-2.0-flash

    importance and personal connection

    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
     Taktlose
    -0.45
     PeEnEo
    -0.43
    PageRoute
    -0.43
    OrNil
    -0.42
     Automatically
    -0.39
     rud
    -0.39
     Easier
    -0.38
     Faster
    -0.37
     hal
    -0.37
     ObjectParameter
    -0.37
    POSITIVE LOGITS
     immensely
    1.19
     greatly
    1.09
     mucho
    1.06
     dearly
    1.05
     tremendously
    1.03
     muchísimo
    1.02
     enemmän
    1.00
     enormously
    0.97
     paljon
    0.97
     much
    0.94
    Activations Density 0.251%

    No Known Activations