Training in progress, step 900, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e045152a73f87855427fb379b7010340bcb5b7bc986cabfaa3a74bd91d905f7
+oid sha256:84586103a211c06663bd404e55c4d764d0a39628f772147e29017239dc0ff834
 size 29034840

last-checkpoint/global_step900/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3162d8665920a7039fb31f5653e692dfa159ad1c3ec79a43a4ad091c587bec42
+size 43429616

last-checkpoint/global_step900/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae6e4291d5589e06115045cc4eeb9a6e9f42c5707e975ee4573111e02947cbfb
+size 43429616

last-checkpoint/global_step900/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3415fbb0b88f14e91742873668a0b376d4c77e6fba3c87c920c0f41835ff0c55
+size 43429616

last-checkpoint/global_step900/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9dd5d41123de36f670814e5f1bba522d4f2c42320f1af5fb37fc1c7e6181839e
+size 43429616

last-checkpoint/global_step900/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7162c4e09532f9490922bccb0dbbcfe539972e0f3d49af32428a76030c053274
+size 637299

last-checkpoint/global_step900/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a1fabfaa7ff216589f1f45e0de64de68782d934b86fc832996ada7c681c3bd9
+size 637171

last-checkpoint/global_step900/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c68864aa3523e8a7232f1228e6e0dba67ac184874fa8332362ced37b3cbe041a
+size 637171

last-checkpoint/global_step900/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:834a420bd57a2a0fe552734d61c0bcc178574ff6e356d461ed6815bc79695c74
+size 637171

last-checkpoint/latest

@@ -1 +1 @@
-global_step850
+global_step900

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36c9044354f826de248840acaaec171f816609c147a664089731a0570deef948
+oid sha256:b8f22ced19e790cc864cefe3b7c711d9ae631c44f95d42fb4829688cc3de0153
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54f4c4013326db4e7267b656aaf72b86570f8aeee91ad39242a416cf8b963191
+oid sha256:3e0407513eba77d34cbf3adf0e59a58bd80716f4f00f414854253637e82be43d
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43e910793831957d8685c316138c33eef8867edf60052477dc9ad6ec0c6da901
+oid sha256:6060636c023258ce9b965e244b8a58b4c99d5784dde4405b39737550ef50cd4f
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7cfd0d71ef5a6c58d9f1d46851f4b1e699ca8a50ab3223cfb39668895cffeef2
+oid sha256:c24ccdfdcde39cb2265c82c50c36ffdfcc670f757aba4bcf4bb0fdc6d1373c4c
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e24eaa8963cd872c048c2c655789d678b8c3fcd1c77ca0d663feee5857d2e34
+oid sha256:feb4015894f59edc29c71bc4938b5d4ab98daad34a38a3d387b308b3b1d4b280
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": 0.0033526704646646976,
   "best_model_checkpoint": "saves/CADICA_qwenvl_direction_scale4/lora/sft/checkpoint-550",
-  "epoch": 0.43780582024208087,
+  "epoch": 0.4635591037857327,
   "eval_steps": 50,
-  "global_step": 850,
+  "global_step": 900,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1520,11 +1520,100 @@
       "eval_steps_per_second": 0.761,
       "num_input_tokens_seen": 8486400,
       "step": 850
+    },
+    {
+      "epoch": 0.44038114859644606,
+      "grad_norm": 0.22602261311014887,
+      "learning_rate": 8.930720915538487e-05,
+      "loss": 0.0255,
+      "num_input_tokens_seen": 8536320,
+      "step": 855
+    },
+    {
+      "epoch": 0.44295647695081125,
+      "grad_norm": 0.1515497727795517,
+      "learning_rate": 8.915646321965614e-05,
+      "loss": 0.0267,
+      "num_input_tokens_seen": 8586240,
+      "step": 860
+    },
+    {
+      "epoch": 0.4455318053051764,
+      "grad_norm": 0.039053785243136956,
+      "learning_rate": 8.900479122779712e-05,
+      "loss": 0.0107,
+      "num_input_tokens_seen": 8636160,
+      "step": 865
+    },
+    {
+      "epoch": 0.4481071336595416,
+      "grad_norm": 0.42882717357803835,
+      "learning_rate": 8.885219676687277e-05,
+      "loss": 0.0277,
+      "num_input_tokens_seen": 8686080,
+      "step": 870
+    },
+    {
+      "epoch": 0.45068246201390677,
+      "grad_norm": 5.519940082555776,
+      "learning_rate": 8.869868344576459e-05,
+      "loss": 0.0175,
+      "num_input_tokens_seen": 8736000,
+      "step": 875
+    },
+    {
+      "epoch": 0.45325779036827196,
+      "grad_norm": 0.15983443395533375,
+      "learning_rate": 8.854425489508532e-05,
+      "loss": 0.0065,
+      "num_input_tokens_seen": 8785920,
+      "step": 880
+    },
+    {
+      "epoch": 0.45583311872263715,
+      "grad_norm": 0.42885948370195226,
+      "learning_rate": 8.838891476709288e-05,
+      "loss": 0.0476,
+      "num_input_tokens_seen": 8835840,
+      "step": 885
+    },
+    {
+      "epoch": 0.4584084470770023,
+      "grad_norm": 0.18209352469474613,
+      "learning_rate": 8.823266673560426e-05,
+      "loss": 0.018,
+      "num_input_tokens_seen": 8885760,
+      "step": 890
+    },
+    {
+      "epoch": 0.4609837754313675,
+      "grad_norm": 1.7895291298260116,
+      "learning_rate": 8.807551449590846e-05,
+      "loss": 0.0337,
+      "num_input_tokens_seen": 8935680,
+      "step": 895
+    },
+    {
+      "epoch": 0.4635591037857327,
+      "grad_norm": 0.11579050055375595,
+      "learning_rate": 8.791746176467907e-05,
+      "loss": 0.0063,
+      "num_input_tokens_seen": 8985600,
+      "step": 900
+    },
+    {
+      "epoch": 0.4635591037857327,
+      "eval_loss": 0.010749292559921741,
+      "eval_runtime": 19.2918,
+      "eval_samples_per_second": 3.11,
+      "eval_steps_per_second": 0.778,
+      "num_input_tokens_seen": 8985600,
+      "step": 900
     }
   ],
   "logging_steps": 5,
   "max_steps": 3400,
-  "num_input_tokens_seen": 8486400,
+  "num_input_tokens_seen": 8985600,
   "num_train_epochs": 2,
   "save_steps": 50,
   "stateful_callbacks": {
@@ -1539,7 +1628,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 559924481097728.0,
+  "total_flos": 592864460668928.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null