Spaces:

jacmor64
/

Ares_v1

Paused

App Files Files Community

Ares_v1 / src /ares /training

13.5 kB

Ctrl+K

Ctrl+K

3 contributors

History: 1 commit

Ares Deployer

Deploy Ares full from scratch: BPE 128K, RoPE 8192, GQA+KV, RMSNorm, SwiGLU, RAG SQLite, CoT/ToT/Planner, SFT/RLHF, code+search

701cf7d 23 days ago

__init__.py

0 Bytes
Deploy Ares full from scratch: BPE 128K, RoPE 8192, GQA+KV, RMSNorm, SwiGLU, RAG SQLite, CoT/ToT/Planner, SFT/RLHF, code+search 23 days ago
dataset_pipeline.py

5.68 kB
Deploy Ares full from scratch: BPE 128K, RoPE 8192, GQA+KV, RMSNorm, SwiGLU, RAG SQLite, CoT/ToT/Planner, SFT/RLHF, code+search 23 days ago
loss.py

1.24 kB
Deploy Ares full from scratch: BPE 128K, RoPE 8192, GQA+KV, RMSNorm, SwiGLU, RAG SQLite, CoT/ToT/Planner, SFT/RLHF, code+search 23 days ago
optimizer.py

1.63 kB
Deploy Ares full from scratch: BPE 128K, RoPE 8192, GQA+KV, RMSNorm, SwiGLU, RAG SQLite, CoT/ToT/Planner, SFT/RLHF, code+search 23 days ago
trainer.py

4.93 kB
Deploy Ares full from scratch: BPE 128K, RoPE 8192, GQA+KV, RMSNorm, SwiGLU, RAG SQLite, CoT/ToT/Planner, SFT/RLHF, code+search 23 days ago