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

    Neuronpedia

    Jacobian LensNEW
    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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 103277
    Prev
    Next
    INDEX
    Explanations

    relationships and comparisons between different entities or concepts

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

    Words or phrases that connect clauses or ideas in a text, particularly comparative constructions, prepositional phrases, transition words, conjunctions, and linguistic relationships between concepts. These include functional elements that establish logical relationships ("than", "or", "as from", "comparable to"), comparative structures ("greater than", "showing greater"), and components of complex phrases that extend meaning.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    balleur
    -0.67
    XmlAccessorType
    -0.57
     kasarigan
    -0.53
    eleste
    -0.51
    pencer
    -0.51
    togram
    -0.50
    vestream
    -0.50
    bukkit
    -0.49
     Stray
    -0.49
     Stool
    -0.49
    POSITIVE LOGITS
     agrav
    0.43
    GEBURTSDATUM
    0.40
     severity
    0.38
    PyExc
    0.37
    Източници
    0.35
     cref
    0.35
    AutoScaleMode
    0.33
     schlimm
    0.32
     threatening
    0.32
    Worse
    0.31
    Activations Density 0.291%

    No Known Activations