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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 16-GEMMASCOPE-RES-16K
    6. 14908
    Prev
    Next
    INDEX
    Explanations

    code statements that indicate the status of a variable or condition

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

    Code snippets or special characters

    np_acts-logits-general · gemini-2.0-flash

    HIV or break

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ocellular
    -0.43
    oopy
    -0.43
    var
    -0.42
    Pas
    -0.42
    పు
    -0.41
     peix
    -0.41
     насељу
    -0.41
     locus
    -0.40
    ophyte
    -0.40
    XmlAccessType
    -0.40
    POSITIVE LOGITS
    break
    1.21
     break
    1.15
     breaks
    0.88
     Break
    0.87
    Break
    0.86
    breaks
    0.82
    脚注の使い方
    0.79
     CreateTagHelper
    0.79
     BREAK
    0.78
    BREAK
    0.77
    Activations Density 0.177%

    No Known Activations