Defeating the Training-Inference Mismatch via FP16
Sea AI Lab
company
Verified
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
Rethinking the Trust Region in LLM Reinforcement Learning
Revisiting Parameter Server in LLM Post-Training
spaces 7
Runtime error
27
Sailor2 20B Chat
π±
Chat with Sailor2 for detailed answers in multiple languages
Running
12
Scaling With Vocab Demo
π
Predict optimal vocabulary size for models
Running
4
Pipeline Parallellism with Controllable Memory
π
Calculate and visualize pipeline schedules
Running
21
Zero Bubble Pipeline Parallellism
π
Calculate and visualize pipeline schedules
Running
6
RegMix
π
Generate predictions and visualize regression results from CSV data
Runtime error
6
Sailor 14B Chat
β
Generate responses to text questions in multiple languages
models 79
sail/longspec-Llama-3-8B-Instruct-262k
Text Generation β’ 0.3B β’ Updated
β’ 1
sail/longspec-QwQ-32B-Preview
Text Generation β’ 0.6B β’ Updated
β’ 11
sail/longspec-vicuna-13b-v1.5-16k
Text Generation β’ 0.4B β’ Updated
β’ 2
sail/longspec-longchat-13b-16k
Text Generation β’ 0.4B β’ Updated
β’ 1
sail/longspec-vicuna-7b-v1.5-16k
Text Generation β’ 0.3B β’ Updated
sail/longspec-longchat-7b-v1.5-32k
Text Generation β’ 0.3B β’ Updated
β’ 2
sail/Qwen2.5-Math-7B-Oat-Zero
Text Generation β’ 8B β’ Updated
β’ 70 β’ β’ 6
sail/Qwen2.5-Math-1.5B-Oat-Zero
Text Generation β’ 2B β’ Updated
β’ 115 β’ β’ 4
sail/Llama-3.2-3B-Oat-Zero
Text Generation β’ 3B β’ Updated
β’ 53 β’ 1
sail/Sailor2-20B
Text Generation β’ 19B β’ Updated
β’ 31 β’ 10
datasets 8
sail/Sanity-Test-R1D-1.5B
Viewer
β’ Updated
β’ 1.52k β’ 43 β’ 7
sail/longspec-data
Preview
β’ Updated
β’ 223 β’ 2
sail/ActPRMData
Viewer
β’ Updated
β’ 663k β’ 78 β’ 1
sail/regmix-data
Viewer
β’ Updated
β’ 13.7M β’ 1.39k β’ 4
sail/regmix-data-sample
Viewer
β’ Updated
β’ 698k β’ 343 β’ 2
sail/Sailcompass_data
Preview
β’ Updated
β’ 14
sail/sailcraft_lm_resource
Updated
β’ 390 β’ 1
sail/symbolic-instruction-tuning
Viewer
β’ Updated
β’ 875k β’ 1.26k β’ 15