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. 62230
    Prev
    Next
    INDEX
    Explanations

    scientific and procedural language markers in a study or report

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

    Words and phrases that signal relationships between ideas, such as transitions, explanations, comparisons, or logical connections in formal or technical text.

    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
    RTEE
    -0.51
     arşivlendi
    -0.46
    aarrggbb
    -0.42
    揄
    -0.41
    AsUp
    -0.40
    +#+#
    -0.39
    GEBURTSDATUM
    -0.39
     pinulongan
    -0.38
    FirstResponder
    -0.38
    ."],
    -0.37
    POSITIVE LOGITS
    RenderAtEndOf
    0.65
     跳转至
    0.55
    Примітки
    0.49
    TagHelper
    0.49
     Administrativna
    0.46
     resourceCulture
    0.44
     MemoryStream
    0.43
     Himo
    0.42
     cremallera
    0.42
    findpost
    0.40
    Activations Density 0.252%

    No Known Activations