github synced commits to refs/pull/811/merge at github/minimind from mirror
-
df7f71c68d Merge 864bd6743749ad3603ab8ca686d5faefc5031388 into 89d674b8a517010f5561b6d8ab2dcbb58e2fb91b
-
89d674b8a5 [refactor] PPO cleanup
-
2d4c385b0f [fix] PPO rollout mode
-
e9f46506da Merge pull request #814 from FloatFrank/fix/ppo-logp-autocast-align-v3
-
1191fc8e9f [fix] PPO: align mb_resp_logp with old_resp_logp under autocast
- Compare 5 commits »
github
synced and deleted reference refs/tags/refs/pull/814/merge at github/minimind from mirror
github synced commits to master at github/minimind from mirror
-
89d674b8a5 [refactor] PPO cleanup
-
2d4c385b0f [fix] PPO rollout mode
-
e9f46506da Merge pull request #814 from FloatFrank/fix/ppo-logp-autocast-align-v3
-
1191fc8e9f [fix] PPO: align mb_resp_logp with old_resp_logp under autocast
- Compare 4 commits »
github synced commits to refs/pull/780/merge at github/minimind from mirror
-
497236be3c Merge 450885c0d5bd1e471f7393abdd96bd605322f49f into 89d674b8a517010f5561b6d8ab2dcbb58e2fb91b
-
89d674b8a5 [refactor] PPO cleanup
-
2d4c385b0f [fix] PPO rollout mode
-
e9f46506da Merge pull request #814 from FloatFrank/fix/ppo-logp-autocast-align-v3
-
1191fc8e9f [fix] PPO: align mb_resp_logp with old_resp_logp under autocast
- Compare 5 commits »
github synced commits to refs/pull/808/merge at github/minimind from mirror
-
fedfbf850d Merge 1e0981d1e371452d254331807406edc129d6e649 into 512eed0b6556e741d80864f054d45d271459772a
-
1e0981d1e3 feat: make code generation runs resumable
-
cf2329fab0 feat: add reproducible MBPP code evaluation
-
8cbbda005c fix: tighten runtime and GRPO metrics
-
6a98206aef docs: keep repository documentation project-only
- Compare 5 commits »
github synced commits to refs/pull/808/head at github/minimind from mirror
-
1e0981d1e3 feat: make code generation runs resumable
-
cf2329fab0 feat: add reproducible MBPP code evaluation
-
8cbbda005c fix: tighten runtime and GRPO metrics
-
6a98206aef docs: keep repository documentation project-only
- Compare 4 commits »
github
synced and deleted reference refs/tags/refs/pull/128/merge at github/minimind-v from mirror
github synced commits to refs/pull/762/merge at github/minimind from mirror
-
0fee6a001a Merge 743420daacc87b979e6db2f440e8a7f02733fc67 into 512eed0b6556e741d80864f054d45d271459772a
-
512eed0b65 [fix] robustness
- Compare 2 commits »
github synced commits to refs/pull/780/head at github/minimind from mirror
-
450885c0d5 feat: 添加 macOS Metal (MPS) 设备支持
-
512eed0b65 [fix] robustness
-
3f1a7cc25b [fix] ddp exit hang
-
4a68da72d5 [fix] lora dims
- Compare 4 commits »
github synced commits to refs/pull/780/merge at github/minimind from mirror
-
66f3450870 Merge 450885c0d5bd1e471f7393abdd96bd605322f49f into 512eed0b6556e741d80864f054d45d271459772a
-
450885c0d5 feat: 添加 macOS Metal (MPS) 设备支持
- Compare 2 commits »
github synced commits to refs/pull/708/merge at github/minimind from mirror
-
946953dbe2 Merge 138c367ddb2ae269ad4c1d81c1b0c62f9ffc5e8e into 512eed0b6556e741d80864f054d45d271459772a
-
512eed0b65 [fix] robustness
- Compare 2 commits »
github
synced and deleted reference refs/tags/refs/pull/803/merge at github/minimind from mirror
github synced commits to refs/pull/803/head at github/minimind from mirror
-
b40eb54434 FSDP TP
-
da93141a0d update ignore
-
21d501f385 profile.
- Compare 3 commits »
github synced commits to refs/pull/803/merge at github/minimind from mirror
-
85069e9e35 Merge b40eb544347eab79c6d224d48c7a4732ceab8bd7 into 512eed0b6556e741d80864f054d45d271459772a
-
b40eb54434 FSDP TP
-
da93141a0d update ignore
-
21d501f385 profile.
- Compare 4 commits »
github synced commits to refs/pull/803/head at github/minimind from mirror
-
4751e359da distributed training.
-
02b61435b1 fix
-
a6a19aa877 update
- Compare 3 commits »
github synced commits to refs/pull/803/merge at github/minimind from mirror
-
3deaff595c Merge 4751e359da1ec00f9e82b7a757d5ca0893ff1cf2 into 512eed0b6556e741d80864f054d45d271459772a
-
4751e359da distributed training.
-
02b61435b1 fix
-
a6a19aa877 update
- Compare 4 commits »
github synced commits to refs/pull/803/head at github/minimind from mirror
-
b59f54cfc5 Add script to download skypile data
-
2ce02ea4ab optimize SFT training speed
- Compare 2 commits »
github synced commits to refs/pull/803/merge at github/minimind from mirror
-
e38f16c912 Merge b59f54cfc594367452197cd636a4493986e22c5f into 512eed0b6556e741d80864f054d45d271459772a
-
b59f54cfc5 Add script to download skypile data
-
2ce02ea4ab optimize SFT training speed
-
17f67f9897 Add MFU cal utils.
- Compare 4 commits »
github synced commits to refs/pull/803/head at github/minimind from mirror
-
17f67f9897 Add MFU cal utils.
github synced commits to refs/pull/784/merge at github/minimind from mirror
-
489d96638e Merge 0761ffa7cda81954f5348e7dcf9b4f7e2fcb4064 into 512eed0b6556e741d80864f054d45d271459772a
-
512eed0b65 [fix] robustness
- Compare 2 commits »