Training in progress, step 1150, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f682c4f26b2d865ab23e7546659c851c99ad1ed58aaaeda4b6412b57ead1e67
+oid sha256:d69e5b496c36ad259db37344eb706b381e79be34b12a26861d7f318295dcfd58
 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 sha256:7ccfa430b965964025c7df588274efc77bdd72607de30b33a82d4ed872ea6608
+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 sha256:43f0b7076dba6248918b48b55a277f7527b813dbd05628c8d8f96c1a6b9ac31c
+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 sha256:6b9d3bb5d31c6296f835ae398f02b670bba2705dc33431cd01123a0c3e3b5819
+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:af802e7d219ff07b9063257ff4308ed252d77ed99f36064c186043c8c6386cb2
+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:a5668efa7a2d3e6e82cce1fb88dafdddf66e36e9f594edbe8b5128192eff0e46
+oid sha256:dc7c1c90680adb84a90114a256d228473c266c4da4454cc4d39e70f828ce64ad
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.4378110468387604,
-  "best_model_checkpoint": "saves/CADICA_qwenvl_direction_then_DetectAndClassify_scale6/lora/sft/checkpoint-1050",
-  "epoch": 0.3249630723781389,
+  "best_metric": 0.42494550347328186,
+  "best_model_checkpoint": "saves/CADICA_qwenvl_direction_then_DetectAndClassify_scale6/lora/sft/checkpoint-1150",
+  "epoch": 0.3397341211225997,
   "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": 0.789,
       "num_input_tokens_seen": 11413120,
       "step": 1100
+    },
+    {
+      "epoch": 0.32644017725258495,
+      "grad_norm": 5.808097944570942,
+      "learning_rate": 9.654007459141634e-05,
+      "loss": 0.4121,
+      "num_input_tokens_seen": 11465064,
+      "step": 1105
+    },
+    {
+      "epoch": 0.32791728212703103,
+      "grad_norm": 4.015479597894709,
+      "learning_rate": 9.649529535836887e-05,
+      "loss": 0.4569,
+      "num_input_tokens_seen": 11516304,
+      "step": 1110
+    },
+    {
+      "epoch": 0.3293943870014771,
+      "grad_norm": 8.768813687811088,
+      "learning_rate": 9.645023873485557e-05,
+      "loss": 0.4121,
+      "num_input_tokens_seen": 11568568,
+      "step": 1115
+    },
+    {
+      "epoch": 0.3308714918759232,
+      "grad_norm": 10.930663245586304,
+      "learning_rate": 9.640490498968383e-05,
+      "loss": 0.4112,
+      "num_input_tokens_seen": 11620672,
+      "step": 1120
+    },
+    {
+      "epoch": 0.33234859675036926,
+      "grad_norm": 6.691827883878219,
+      "learning_rate": 9.63592943933143e-05,
+      "loss": 0.3564,
+      "num_input_tokens_seen": 11672864,
+      "step": 1125
+    },
+    {
+      "epoch": 0.33382570162481534,
+      "grad_norm": 8.33100451031768,
+      "learning_rate": 9.631340721785934e-05,
+      "loss": 0.3909,
+      "num_input_tokens_seen": 11724128,
+      "step": 1130
+    },
+    {
+      "epoch": 0.3353028064992615,
+      "grad_norm": 7.005971082198048,
+      "learning_rate": 9.62672437370814e-05,
+      "loss": 0.4636,
+      "num_input_tokens_seen": 11776416,
+      "step": 1135
+    },
+    {
+      "epoch": 0.33677991137370755,
+      "grad_norm": 14.154463913713748,
+      "learning_rate": 9.622080422639133e-05,
+      "loss": 0.4617,
+      "num_input_tokens_seen": 11828256,
+      "step": 1140
+    },
+    {
+      "epoch": 0.33825701624815363,
+      "grad_norm": 14.751813027169304,
+      "learning_rate": 9.617408896284678e-05,
+      "loss": 0.3443,
+      "num_input_tokens_seen": 11882048,
+      "step": 1145
+    },
+    {
+      "epoch": 0.3397341211225997,
+      "grad_norm": 3.2576085972339706,
+      "learning_rate": 9.612709822515054e-05,
+      "loss": 0.4432,
+      "num_input_tokens_seen": 11933632,
+      "step": 1150
+    },
+    {
+      "epoch": 0.3397341211225997,
+      "eval_loss": 0.42494550347328186,
+      "eval_runtime": 19.6038,
+      "eval_samples_per_second": 3.061,
+      "eval_steps_per_second": 0.765,
+      "num_input_tokens_seen": 11933632,
+      "step": 1150
     }
   ],
   "logging_steps": 5,
   "max_steps": 6770,
-  "num_input_tokens_seen": 11413120,
+  "num_input_tokens_seen": 11933632,
   "num_train_epochs": 2,
   "save_steps": 50,
   "stateful_callbacks": {
@@ -1984,7 +2073,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 752909914210304.0,
+  "total_flos": 787259637891072.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null