© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 35-GEMMASCOPE-RES-16K
    6. 14285
    Prev
    Next
    INDEX
    Explanations

    structural focus in scientific writing

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.35.hook_resid_post
    Hook Layer
    35
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    intenant
    -0.44
    zzleHttp
    -0.44
    eseorang
    -0.42
    UKAN
    -0.42
    MenuInflater
    -0.41
    liferay
    -0.41
     ואת
    -0.41
     Remember
    -0.40
    execSQL
    -0.40
     tarvit
    -0.39
    POSITIVE LOGITS
    Abstract
    1.23
     Abstract
    1.21
    ABSTRACT
    1.09
     abstract
    1.08
     ABSTRACT
    1.07
    abstract
    0.98
     abstracts
    0.81
    Keywords
    0.81
     abstra
    0.81
     Aims
    0.80
    Activations Density 1.675%

    No Known Activations