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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 2219
    Prev
    Next
    INDEX
    Explanations

    elements related to decision-making and the consequences of actions in societal contexts

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

    followed by imperative verbs

    np_acts-logits-general · gemini-2.0-flash

    actions like giving, getting, using

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    ']).
    -0.49
    ']}
    -0.49
    }}],
    -0.48
     NgModule
    -0.48
     semn
    -0.47
    '])
    -0.47
    ')]
    -0.47
     känd
    -0.46
     confirmation
    -0.45
     vero
    -0.45
    POSITIVE LOGITS
     befri
    0.86
     hired
    0.80
     masturb
    0.80
    mybatisplus
    0.75
     wear
    0.75
     donate
    0.73
     hiring
    0.73
     donated
    0.73
     become
    0.73
     resourceCulture
    0.73
    Activations Density 0.975%

    No Known Activations