Training in progress, step 1150, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f16f6078244d0994120e51d527f92194dad6fa70876b55e40b787da46626e91a
+oid sha256:050dc1f1854d68a42fb308d9dd143ddc3d6fc0d86870ea1eac6bf90317df973a
 size 29034840

last-checkpoint/global_step1150/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:e55a1500f8b0709fe979be0a0d2ee80da98483414a2d9cb6c031c4973ee192ef
+size 43429616

last-checkpoint/global_step1150/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:e0caf7ee7bd7ce7e7503ae05524992763bb3bb77347dbe068bcae95b94b3fcc2
+size 43429616

last-checkpoint/global_step1150/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58dcc807ed02b7212af3f3181a1663fa62c42f0666df545a96e79db27d9d71e7
+size 43429616

last-checkpoint/global_step1150/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d4070d3a090f246cdf0d8dc238f5a3d89f86e7b8eeba2f18d6facbb9d474005
+size 43429616

last-checkpoint/global_step1150/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa758cf6fa59b0c76382480e4cb3a02f4e3bcc7c5814e40e5eff7203544b973f
+size 637299

last-checkpoint/global_step1150/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a739833690f638f8f3ea52ef250f6914f0d6cc7cf09d4ba85216fa45aed5b895
+size 637171

last-checkpoint/global_step1150/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64a9559916dc6f652d640305c67706ef9b17164de4b12a7f651920cf80c4d653
+size 637171

last-checkpoint/global_step1150/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94a93cc4afb60025355716cc53e16bf4936ae88fb37858092e9cfc5c70e1b9a9
+size 637171

last-checkpoint/latest

@@ -1 +1 @@
-global_step1100
+global_step1150

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d9fea52fb92cc51e76feeb2b139ce35723c0cb651da383e4f7eec2606ed6c2a
+oid sha256:274dc3860ee0c7f4d5348f60910a4b568498c04adfefb89f905b1c78a82c1312
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1b5820ebfcc2e1cfe1ad2619a05ea9a484ff21635e13e386bf14abd302f2c0f7
+oid sha256:e9aa441491b9ca89e796944520fa1db332a67c0a1a920be83edd2d96d741716d
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb8d96a68e732fca41980516622a50990bbd3ee989e72076a35c8608d9b4d136
+oid sha256:c1ee3434533b24fb771504fa8cceb5c2ea25fe0de1641128feaceccc65afe6ed
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce6f1db6d09f4d89a9b2bd8dc8eeb99f1fada2ec04376e23b5a7a13004994005
+oid sha256:93b4a44be1335173d2e3120bd0d1e6346f3e832d8935752c70ce1e98f017fa87
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d0c7456eafeee3179566bb381c9153771d7e0f21738d2398944d053915d0651
+oid sha256:7d461c8d7517d4b88333bff7984fc3bfc149292198b04bbc18a49aee698ffb5c
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.0003656313638202846,
-  "best_model_checkpoint": "saves/CADICA_qwenvl_direction_scale4/lora/sft/checkpoint-1100",
-  "epoch": 0.56657223796034,
+  "best_metric": 1.5566551155643538e-05,
+  "best_model_checkpoint": "saves/CADICA_qwenvl_direction_scale4/lora/sft/checkpoint-1150",
+  "epoch": 0.5923255215039918,
   "eval_steps": 50,
-  "global_step": 1100,
+  "global_step": 1150,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1965,11 +1965,100 @@
       "eval_steps_per_second": 0.777,
       "num_input_tokens_seen": 10982400,
       "step": 1100
+    },
+    {
+      "epoch": 0.5691475663147051,
+      "grad_norm": 0.020540774848059624,
+      "learning_rate": 8.07106356344834e-05,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 11032320,
+      "step": 1105
+    },
+    {
+      "epoch": 0.5717228946690703,
+      "grad_norm": 0.049646985711380834,
+      "learning_rate": 8.051838793910038e-05,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 11082240,
+      "step": 1110
+    },
+    {
+      "epoch": 0.5742982230234355,
+      "grad_norm": 0.02410302194894019,
+      "learning_rate": 8.032541847934146e-05,
+      "loss": 0.001,
+      "num_input_tokens_seen": 11132160,
+      "step": 1115
+    },
+    {
+      "epoch": 0.5768735513778007,
+      "grad_norm": 0.03404095257940449,
+      "learning_rate": 8.013173181896283e-05,
+      "loss": 0.0219,
+      "num_input_tokens_seen": 11182080,
+      "step": 1120
+    },
+    {
+      "epoch": 0.5794488797321659,
+      "grad_norm": 0.0005611895912019751,
+      "learning_rate": 7.993733253868256e-05,
+      "loss": 0.0,
+      "num_input_tokens_seen": 11232000,
+      "step": 1125
+    },
+    {
+      "epoch": 0.582024208086531,
+      "grad_norm": 0.03678453984585236,
+      "learning_rate": 7.974222523607236e-05,
+      "loss": 0.0,
+      "num_input_tokens_seen": 11281920,
+      "step": 1130
+    },
+    {
+      "epoch": 0.5845995364408962,
+      "grad_norm": 0.002820815423322981,
+      "learning_rate": 7.954641452544865e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 11331840,
+      "step": 1135
+    },
+    {
+      "epoch": 0.5871748647952614,
+      "grad_norm": 0.003003123169716893,
+      "learning_rate": 7.934990503776363e-05,
+      "loss": 0.0,
+      "num_input_tokens_seen": 11381760,
+      "step": 1140
+    },
+    {
+      "epoch": 0.5897501931496266,
+      "grad_norm": 0.020018776153065302,
+      "learning_rate": 7.915270142049566e-05,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 11431680,
+      "step": 1145
+    },
+    {
+      "epoch": 0.5923255215039918,
+      "grad_norm": 0.0005407295497242103,
+      "learning_rate": 7.89548083375394e-05,
+      "loss": 0.0127,
+      "num_input_tokens_seen": 11481600,
+      "step": 1150
+    },
+    {
+      "epoch": 0.5923255215039918,
+      "eval_loss": 1.5566551155643538e-05,
+      "eval_runtime": 19.3425,
+      "eval_samples_per_second": 3.102,
+      "eval_steps_per_second": 0.775,
+      "num_input_tokens_seen": 11481600,
+      "step": 1150
     }
   ],
   "logging_steps": 5,
   "max_steps": 3400,
-  "num_input_tokens_seen": 10982400,
+  "num_input_tokens_seen": 11481600,
   "num_train_epochs": 2,
   "save_steps": 50,
   "stateful_callbacks": {
@@ -1984,7 +2073,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 724624378953728.0,
+  "total_flos": 757564358524928.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null