Training in progress, step 150, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54617ace9be9da2e4e57ae55a0f5329011112b5c8868c9fffa532ea69e4c2d98
+oid sha256:e1db5746fd65250f2bf6817d2de9266703177b7d0c3decef80aefa9773e3ca7a
 size 29034840

last-checkpoint/global_step150/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:e6af200421fe1d73ecd1d42ad1ed20ccf899c9f21305ebaefb140aada662a759
+size 43429616

last-checkpoint/global_step150/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:be78431879fe26d398ef6cc8424c45ea0c0d43e6486b3290d03d1a823c1d996f
+size 43429616

last-checkpoint/global_step150/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61ebcbde783ef855f8a86fcbaaef684c07b0acc7cc73acc0425ea5522428c95d
+size 43429616

last-checkpoint/global_step150/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:eb9a0b3a0e13aef6722af4cbff4275803aab00e1349f0db43f5927fc6d027d43
+size 43429616

last-checkpoint/global_step150/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34f144f33b7ac763950c684990157dad4d1ee523ae43797b9a6b90a998a9d314
+size 637299

last-checkpoint/global_step150/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3e0249ea361e86670960dca0ca756a903dbb19568eef1a99e26ae12785a1d69
+size 637171

last-checkpoint/global_step150/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82afbf3944b11f68d9499cbaa773c1fcc29e1a0e9b4eb333a671ffa88a5222fe
+size 637171

last-checkpoint/global_step150/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a0caccf9e31a439cb573c5600a0e4625f1ff654d897f43ccfa9da9d8591fe0e
+size 637171

last-checkpoint/latest

@@ -1 +1 @@
-global_step100
+global_step150

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:308f94f9a5c24e1bad5c393d56ae7af7782600f4e791d9c6ac35b22fff2105b6
+oid sha256:70cc56408014c410353d4dd58ae9b03f4be043f5f800324f66fd8e20e99b840e
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b056f3c23cb32dc77a2ec9e7651e0b64e4440e21f0fdf969b86bfc56a1cbdf06
+oid sha256:49d1438e98cc9c53a6852464635ce62e9788e61eb3646b73e33813f487c4b6ae
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f3f8a05714bc528f4885a2816181652f2303b3e8150f89b56aaee6bec56aa520
+oid sha256:4388add9cec90932f8ff0100d27a0574d98e1bad52ff89d44e31967d2b4fbfde
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f755bd3c330281961e5c03af9d10ce8c1e1678619d384f6f1fd5fd7dce2ff50
+oid sha256:a705d6dfaae4f2c1b4b2be6b25a6eb521ffae6fcba21cc1531e97b60037ed079
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4e451a9e086b06d7c667be8442b2115f5c088953bade0b625e61f2ce5c7fd404
+oid sha256:c587ef55825cdfdbac47c29c3bd8a2996263c16833a5bd6aaa2fb014bea1e9d1
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.18663176894187927,
-  "best_model_checkpoint": "saves/CADICA_qwenvl_direction_scale4/lora/sft/checkpoint-100",
-  "epoch": 0.05150656708730363,
+  "best_metric": 0.09665286540985107,
+  "best_model_checkpoint": "saves/CADICA_qwenvl_direction_scale4/lora/sft/checkpoint-150",
+  "epoch": 0.07725985063095545,
   "eval_steps": 50,
-  "global_step": 100,
+  "global_step": 150,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -185,11 +185,100 @@
       "eval_steps_per_second": 0.793,
       "num_input_tokens_seen": 998400,
       "step": 100
+    },
+    {
+      "epoch": 0.05408189544166881,
+      "grad_norm": 2.2181531422996716,
+      "learning_rate": 6.176470588235295e-05,
+      "loss": 0.168,
+      "num_input_tokens_seen": 1048320,
+      "step": 105
+    },
+    {
+      "epoch": 0.056657223796033995,
+      "grad_norm": 3.1829920225573236,
+      "learning_rate": 6.470588235294118e-05,
+      "loss": 0.0709,
+      "num_input_tokens_seen": 1098240,
+      "step": 110
+    },
+    {
+      "epoch": 0.05923255215039917,
+      "grad_norm": 4.337350477588576,
+      "learning_rate": 6.764705882352942e-05,
+      "loss": 0.1609,
+      "num_input_tokens_seen": 1148160,
+      "step": 115
+    },
+    {
+      "epoch": 0.06180788050476436,
+      "grad_norm": 2.1010046045637365,
+      "learning_rate": 7.058823529411765e-05,
+      "loss": 0.0354,
+      "num_input_tokens_seen": 1198080,
+      "step": 120
+    },
+    {
+      "epoch": 0.06438320885912954,
+      "grad_norm": 2.232308844812103,
+      "learning_rate": 7.352941176470589e-05,
+      "loss": 0.1133,
+      "num_input_tokens_seen": 1248000,
+      "step": 125
+    },
+    {
+      "epoch": 0.06695853721349472,
+      "grad_norm": 5.641631090993415,
+      "learning_rate": 7.647058823529411e-05,
+      "loss": 0.0867,
+      "num_input_tokens_seen": 1297920,
+      "step": 130
+    },
+    {
+      "epoch": 0.0695338655678599,
+      "grad_norm": 1.5031437609685787,
+      "learning_rate": 7.941176470588235e-05,
+      "loss": 0.1352,
+      "num_input_tokens_seen": 1347840,
+      "step": 135
+    },
+    {
+      "epoch": 0.07210919392222509,
+      "grad_norm": 3.2992644431188465,
+      "learning_rate": 8.23529411764706e-05,
+      "loss": 0.101,
+      "num_input_tokens_seen": 1397760,
+      "step": 140
+    },
+    {
+      "epoch": 0.07468452227659027,
+      "grad_norm": 3.494236832758233,
+      "learning_rate": 8.529411764705883e-05,
+      "loss": 0.0334,
+      "num_input_tokens_seen": 1447680,
+      "step": 145
+    },
+    {
+      "epoch": 0.07725985063095545,
+      "grad_norm": 0.0602113869322109,
+      "learning_rate": 8.823529411764706e-05,
+      "loss": 0.0667,
+      "num_input_tokens_seen": 1497600,
+      "step": 150
+    },
+    {
+      "epoch": 0.07725985063095545,
+      "eval_loss": 0.09665286540985107,
+      "eval_runtime": 19.2745,
+      "eval_samples_per_second": 3.113,
+      "eval_steps_per_second": 0.778,
+      "num_input_tokens_seen": 1497600,
+      "step": 150
     }
   ],
   "logging_steps": 5,
   "max_steps": 3400,
-  "num_input_tokens_seen": 998400,
+  "num_input_tokens_seen": 1497600,
   "num_train_epochs": 2,
   "save_steps": 50,
   "stateful_callbacks": {
@@ -204,7 +293,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 65824787529728.0,
+  "total_flos": 98764767100928.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null