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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 31-GEMMASCOPE-RES-16K
    11. 13420
    Prev
    Next
    INDEX
    Explanations

    phrases indicating exceptions or qualifications in statements

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

    the phrase "is not to say" and similar hedging expressions that introduce contrasting or qualifying statements.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     EconPapers
    -0.48
    RTLR
    -0.44
     NgModule
    -0.41
     bParam
    -0.39
    Atoi
    -0.38
     Throwable
    -0.36
    wieś
    -0.36
    WithMany
    -0.36
     typelib
    -0.36
    ANNELS
    -0.35
    POSITIVE LOGITS
     saying
    1.40
     say
    1.39
     decir
    1.20
    say
    1.20
    Say
    1.19
    saying
    1.18
    Saying
    1.13
     sayin
    1.13
     сказать
    1.10
     SAY
    1.08
    Activations Density 0.236%

    No Known Activations