© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 8402
    Prev
    Next
    INDEX
    Explanations

    terms related to perturbation and its effects

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

    scientific terminology related to perturbation theory or concepts of disruption.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    the word "perturb" and its variations (perturbed, perturbative, perturbation) in technical or scientific contexts.

    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
    󠁴
    -0.43
     TextAlign
    -0.38
    hoeddwyd
    -0.36
    Personensuche
    -0.36
    oneofs
    -0.35
     ujednoznacz
    -0.35
     AspNetCore
    -0.35
     Roskov
    -0.35
    MessageTagHelper
    -0.34
    RTLU
    -0.34
    POSITIVE LOGITS
     interruption
    2.11
     disturbance
    2.09
     disruption
    2.05
     interruptions
    2.05
     interrupting
    1.98
     disturb
    1.98
     interrupt
    1.95
     disruptions
    1.92
     disrupt
    1.91
     interrupted
    1.90
    Activations Density 0.599%

    No Known Activations