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. 18-GEMMASCOPE-RES-16K
    6. 3204
    Prev
    Next
    INDEX
    Explanations

    specific consumer products and experiences

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

    Followed by prepositions or conjunctions

    np_acts-logits-general · gemini-2.0-flash

    compound noun endings

    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
    
    -0.69
    ])).
    -0.66
     proprement
    -0.66
    期刊论文
    -0.65
    ]),
    
    -0.65
     fubject
    -0.62
    ervant
    -0.59
    isContained
    -0.57
    hésitez
    -0.56
     purpoſe
    -0.56
    POSITIVE LOGITS
     fucking
    0.48
     lmao
    0.48
     '{@
    0.47
     everyday
    0.47
    Ữ
    0.47
     bikin
    0.46
     thrown
    0.46
     every
    0.45
     hver
    0.45
     dripping
    0.45
    Activations Density 0.366%

    No Known Activations