© 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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 114869
    Prev
    Next
    INDEX
    Explanations

    instances of the word "join" and its variations, indicating a focus on collaboration or becoming part of a group or initiative

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

    joining groups or people

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    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
     UnityEngine
    -0.43
     extrema
    -0.40
     kró
    -0.40
    nocache
    -0.38
     Negra
    -0.36
     Abstracts
    -0.36
    ています
    -0.36
    innerText
    -0.36
    ')")
    -0.35
     McFar
    -0.34
    POSITIVE LOGITS
    join
    1.26
     join
    1.25
     Join
    1.18
     JOIN
    1.16
    Join
    1.15
     joining
    1.09
     Joining
    1.08
    Joining
    1.07
    joining
    1.05
    joined
    1.05
    Activations Density 0.016%

    No Known Activations