© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 278
    Prev
    Next
    INDEX
    Explanations

    email formatting and headers.

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

    strings of technical codes, email addresses, and a mix of various text formats.

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    the beginning of text or document sections, particularly indicated by tokens like "<bos>" (beginning of sequence) or section headers.

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @bot

    Text after "Q:" or "Search"

    np_acts-logits-general · gemini-2.0-flash

    Japanese katakana transliterations: リー, パー

    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
     from
    -0.56
    as
    -0.51
     sam
    -0.51
     
    -0.50
    at
    -0.50
    <eos>
    -0.49
     are
    -0.49
     in
    -0.49
     as
    -0.49
    스터
    -0.48
    POSITIVE LOGITS
     يتيمه
    1.28
    ThroughAttribute
    1.18
    ніципалі
    1.10
     мәкал
    1.05
    AndEndTag
    1.03
     كومونز
    0.97
     незавершена
    0.96
     utafitiHapana
    0.95
    ChildScrollView
    0.94
    脚注の使い方
    0.94
    Activations Density 0.024%

    No Known Activations