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. 105135
    Prev
    Next
    INDEX
    Explanations

    phrases related to technology and privacy concerns

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

    Text segments marked between special characters (<<>>) represent user comments or responses on online forums like Hacker News, typically replies to other comments that contain discussion about technology, privacy, security issues, or other forum topics.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    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
    Personendaten
    -1.01
    parsedMessage
    -0.97
    GEBURTSDATUM
    -0.96
     queſta
    -0.94
     kasarigan
    -0.92
    AddTagHelper
    -0.85
    rungsseite
    -0.85
     незавершена
    -0.84
     мәкал
    -0.84
    MLLoader
    -0.84
    POSITIVE LOGITS
    response
    0.33
     response
    0.31
    Response
    0.28
      
    0.28
     Response
    0.25
    from
    0.24
     respond
    0.23
    *
    0.22
    0
    0.22
    J
    0.22
    Activations Density 0.109%

    No Known Activations