minimind/trainer
baihaochen 2201060169 fix: switch DeepSpeed config from bf16 to fp16, ZeRO-2 to ZeRO-1 (RTX 3090 compat) 2026-05-23 16:45:56 +08:00
..
ds_config_pretrain.json fix: switch DeepSpeed config from bf16 to fp16, ZeRO-2 to ZeRO-1 (RTX 3090 compat) 2026-05-23 16:45:56 +08:00
rollout_engine.py [fix] inference bug 2026-05-03 20:48:48 +08:00
train_agent.py [feat] add MLA with decoupled RoPE (DeepSeek-V2 style) 2026-05-12 18:38:47 +08:00
train_distillation.py [feat] add MLA with decoupled RoPE (DeepSeek-V2 style) 2026-05-12 18:38:47 +08:00
train_dpo.py [feat] add MLA with decoupled RoPE (DeepSeek-V2 style) 2026-05-12 18:38:47 +08:00
train_full_sft.py fix: add missing --use_mla to train_full_sft.py 2026-05-14 14:28:11 +08:00
train_grpo.py [feat] add MLA with decoupled RoPE (DeepSeek-V2 style) 2026-05-12 18:38:47 +08:00
train_lora.py [feat] add MLA with decoupled RoPE (DeepSeek-V2 style) 2026-05-12 18:38:47 +08:00
train_ppo.py [feat] add MLA with decoupled RoPE (DeepSeek-V2 style) 2026-05-12 18:38:47 +08:00
train_pretrain.py merge local changes 2026-05-12 11:18:45 +00:00
train_pretrain_deepspeed.py chore: remove unused import 2026-05-23 16:42:02 +08:00
train_tokenizer.py merge local changes 2026-05-12 11:18:45 +00:00
trainer_utils.py [feat] add MLA with decoupled RoPE (DeepSeek-V2 style) 2026-05-12 18:38:47 +08:00