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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 1072
    Prev
    Next
    INDEX
    Explanations

    expressions related to technical issues and software performance

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

    Prepositions

    np_acts-logits-general · gemini-2.0-flash

    prepositions before nouns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     pleaſure
    -1.26
     itſelf
    -1.24
     raiſ
    -1.24
     houſe
    -1.20
     Efq
    -1.18
     Majefty
    -1.13
     iſt
    -1.12
     Jefus
    -1.12
     ſtate
    -1.10
     ་་
    -1.09
    POSITIVE LOGITS
     the
    0.90
     them
    0.82
     my
    0.81
     whatever
    0.79
     I
    0.77
     something
    0.76
     mấy
    0.76
     B
    0.73
     @
    0.72
     those
    0.71
    Activations Density 1.071%

    No Known Activations