Training in progress, step 850, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f6287ca2ed8e9ca31735e5d987ece6f9b430689719f525a3a04c74ff709108ab
+oid sha256:6e045152a73f87855427fb379b7010340bcb5b7bc986cabfaa3a74bd91d905f7
 size 29034840

last-checkpoint/global_step850/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:aed5addf75b4dcc2d56305d0370876497c8dabfb4d48e695446fce476aec812b
+size 43429616

last-checkpoint/global_step850/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:c1f82ff3ba505822d500c54af588282805a60dd32165278110e76128b42f988a
+size 43429616

last-checkpoint/global_step850/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ff39496f99b859b31fae012b8e98000809ecb18114c736d7b6939b249d47cb8
+size 43429616

last-checkpoint/global_step850/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ec4187a54aae0ea5e3a79e5b23f9839484130e81587192b2668b5b350aa889f
+size 43429616

last-checkpoint/global_step850/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b18337b921f0aba380566f71c4857b67981734dc7975b4ed43e57711a412ef28
+size 637299

last-checkpoint/global_step850/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d144742df08d3730450d7fb36709b9a81b06278b4025a5f5b7cf8306696ee8ed
+size 637171

last-checkpoint/global_step850/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a1e27c63d553ea81d8a5f4f57f82e6fd2dab396822dedba339d93c3e67b54fd
+size 637171

last-checkpoint/global_step850/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4777d80a9e8c2d0c23709b7630c4e2ffc82a5624bbdac8e3a2d838a612e80caa
+size 637171

last-checkpoint/latest

@@ -1 +1 @@
-global_step800
+global_step850

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b4f003069486a57c6ac033f30cf4c4213eb6b7d659bab68a5a50fdb8da7c4118
+oid sha256:36c9044354f826de248840acaaec171f816609c147a664089731a0570deef948
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a016ef89b4392d083b2c15a7cf06a39bc61a759f648cf6dc03f1c32b89a526aa
+oid sha256:54f4c4013326db4e7267b656aaf72b86570f8aeee91ad39242a416cf8b963191
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9b56fe0893036dc052d18d90feba4328b90ea71561942150b07406ac3d7a700e
+oid sha256:43e910793831957d8685c316138c33eef8867edf60052477dc9ad6ec0c6da901
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0c203d12c2c308dab785ed672c9ca27fb6a2f72acd1e1552d1516c7b0006013
+oid sha256:7cfd0d71ef5a6c58d9f1d46851f4b1e699ca8a50ab3223cfb39668895cffeef2
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5d85b710a5709549c0b4daddcc052f2ed242a5d916ac9ca030c805e7ff501c88
+oid sha256:3e24eaa8963cd872c048c2c655789d678b8c3fcd1c77ca0d663feee5857d2e34
 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": 0.41205253669842906,
+  "epoch": 0.43780582024208087,
   "eval_steps": 50,
-  "global_step": 800,
+  "global_step": 850,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1431,11 +1431,100 @@
       "eval_steps_per_second": 0.776,
       "num_input_tokens_seen": 7987200,
       "step": 800
+    },
+    {
+      "epoch": 0.41462786505279425,
+      "grad_norm": 0.016984614729716928,
+      "learning_rate": 9.076296203830579e-05,
+      "loss": 0.0025,
+      "num_input_tokens_seen": 8037120,
+      "step": 805
+    },
+    {
+      "epoch": 0.4172031934071594,
+      "grad_norm": 8.87156169406028,
+      "learning_rate": 9.062166909283062e-05,
+      "loss": 0.0338,
+      "num_input_tokens_seen": 8087040,
+      "step": 810
+    },
+    {
+      "epoch": 0.4197785217615246,
+      "grad_norm": 0.07123492673307173,
+      "learning_rate": 9.047941543889014e-05,
+      "loss": 0.0505,
+      "num_input_tokens_seen": 8136960,
+      "step": 815
+    },
+    {
+      "epoch": 0.4223538501158898,
+      "grad_norm": 0.3486611879887143,
+      "learning_rate": 9.033620444080428e-05,
+      "loss": 0.0335,
+      "num_input_tokens_seen": 8186880,
+      "step": 820
+    },
+    {
+      "epoch": 0.42492917847025496,
+      "grad_norm": 0.2658900406741178,
+      "learning_rate": 9.019203948553422e-05,
+      "loss": 0.0196,
+      "num_input_tokens_seen": 8236800,
+      "step": 825
+    },
+    {
+      "epoch": 0.42750450682462016,
+      "grad_norm": 0.15327530188348007,
+      "learning_rate": 9.004692398260244e-05,
+      "loss": 0.018,
+      "num_input_tokens_seen": 8286720,
+      "step": 830
+    },
+    {
+      "epoch": 0.43007983517898535,
+      "grad_norm": 5.292901102191953,
+      "learning_rate": 8.9900861364012e-05,
+      "loss": 0.012,
+      "num_input_tokens_seen": 8336640,
+      "step": 835
+    },
+    {
+      "epoch": 0.4326551635333505,
+      "grad_norm": 3.2113529029102375,
+      "learning_rate": 8.975385508416532e-05,
+      "loss": 0.0062,
+      "num_input_tokens_seen": 8386560,
+      "step": 840
+    },
+    {
+      "epoch": 0.4352304918877157,
+      "grad_norm": 0.3685880212421972,
+      "learning_rate": 8.960590861978265e-05,
+      "loss": 0.0232,
+      "num_input_tokens_seen": 8436480,
+      "step": 845
+    },
+    {
+      "epoch": 0.43780582024208087,
+      "grad_norm": 1.4858234531184304,
+      "learning_rate": 8.945702546981969e-05,
+      "loss": 0.0306,
+      "num_input_tokens_seen": 8486400,
+      "step": 850
+    },
+    {
+      "epoch": 0.43780582024208087,
+      "eval_loss": 0.007198736071586609,
+      "eval_runtime": 19.7019,
+      "eval_samples_per_second": 3.045,
+      "eval_steps_per_second": 0.761,
+      "num_input_tokens_seen": 8486400,
+      "step": 850
     }
   ],
   "logging_steps": 5,
   "max_steps": 3400,
-  "num_input_tokens_seen": 7987200,
+  "num_input_tokens_seen": 8486400,
   "num_train_epochs": 2,
   "save_steps": 50,
   "stateful_callbacks": {
@@ -1450,7 +1539,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 526984501526528.0,
+  "total_flos": 559924481097728.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null