Training in progress, step 650, checkpoint

last-checkpoint/adapter_config.json

@@ -19,7 +19,7 @@
   "r": 8,
   "rank_pattern": {},
   "revision": null,
-  "target_modules": "^(?!.*patch_embed).*(?:down_proj|q_proj|fc2|o_proj|gate_proj|up_proj|k_proj|fc1|v_proj|qkv|proj).*",
+  "target_modules": "^(?!.*patch_embed).*(?:up_proj|proj|fc2|fc1|qkv|o_proj|k_proj|gate_proj|v_proj|down_proj|q_proj).*",
   "task_type": "CAUSAL_LM",
   "use_dora": false,
   "use_rslora": false

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f3607f7be0eeae44a002686249cfd8440e2093656b0084bace7dafe752090325
+oid sha256:a040fdacc28b9de2977636cd1c1956bffb7f97df44e5bd2640662f6a98fa674b
 size 29034840

last-checkpoint/global_step650/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:d38ea936450c0cdde6dfdfd8c877bad13d30215a768bfb9d4db61232d766ba53
+size 43429616

last-checkpoint/global_step650/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:d00a58f420ce44852dde4e0d1724f02377fdbe090859b41ed6eaf672a3deea48
+size 43429616

last-checkpoint/global_step650/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2839fddbe80dd74b71e29f35cf804ba105c77ad6fb7db1262c16ead07f282a73
+size 43429616

last-checkpoint/global_step650/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:de1ac5430078e2adeec886f04da5b2323e6bdf488c2942cde8fd74eeb39be709
+size 43429616

last-checkpoint/global_step650/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f9450a0ced61d6c33a66cdf1b2511022f0ec24416b17e1f16ed49dc98ff62dd
+size 637299

last-checkpoint/global_step650/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d01b9093c4ff061c6453535b14beda5a9d96a531472a58effac685152d63de95
+size 637171

last-checkpoint/global_step650/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a3bfb0ef98b44969e473f6d5c11a92b5a607c48dc2a278b5d72591611de4079
+size 637171

last-checkpoint/global_step650/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a41c42f45b9260afbaae285e2a90de1413db2108ea931952bc69942ea6f19e1e
+size 637171

last-checkpoint/latest

@@ -1 +1 @@
-global_step600
+global_step650

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a81e3916b1392c4c49afb171dee5415c15f5a5a5af8749b28195fcfa0596699c
+oid sha256:8044e4c53158c210a17648ba8f2dc2d25a25bbfc55f686015542618eb652a33e
 size 15024

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a781038dd714b87b8adb1aac8dbc8217ceb607428a992133954ad522365236e
+oid sha256:4cd85d7fa425e7888c973f1c2985ac15ca21b5e6171fe140a401c2bc75ca46ff
 size 15024

last-checkpoint/rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9446c3db15f382a5546f13622787fc99392a5e0bc8a9ca2da1838de7ab621a37
+oid sha256:d7915667371a58f1598639e0d1c20a0c59c783c14580cd040a6631eb4ea2311e
 size 15024

last-checkpoint/rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1f11e7a6b3faa884fc23044e3772ff9dd72c257f02e121665061e2a03d518bd9
+oid sha256:35dd78929ad7f0fbf37fdb1284e8edf0424350f6e6ce1cd5a3ee78979af3d3cb
 size 15024

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b76b388bede074656df32b92902ac42b965557bfee0c930366af07d8382b1b4a
+oid sha256:8cd94ecf5c982ee0e060d3e07a575ce03dc3b0f289b5e32a1f65d3b6366a8a0e
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
-  "best_metric": 0.5832681059837341,
-  "best_model_checkpoint": "saves/CADICA_qwenvl_stenosis_detect_scale4/lora/sft/checkpoint-550",
-  "epoch": 0.1545197012619109,
+  "best_metric": 0.5542036890983582,
+  "best_model_checkpoint": "saves/CADICA_qwenvl_stenosis_detect_scale4/lora/sft/checkpoint-650",
+  "epoch": 0.1673963430337368,
   "eval_steps": 50,
-  "global_step": 600,
+  "global_step": 650,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1075,11 +1075,100 @@
       "eval_steps_per_second": 0.783,
       "num_input_tokens_seen": 6299656,
       "step": 600
+    },
+    {
+      "epoch": 0.1558073654390935,
+      "grad_norm": 5.90634634073773,
+      "learning_rate": 9.55911532374151e-05,
+      "loss": 0.6106,
+      "num_input_tokens_seen": 6351680,
+      "step": 605
+    },
+    {
+      "epoch": 0.15709502961627608,
+      "grad_norm": 3.5429043559071034,
+      "learning_rate": 9.549077815930636e-05,
+      "loss": 0.5812,
+      "num_input_tokens_seen": 6403648,
+      "step": 610
+    },
+    {
+      "epoch": 0.15838269379345868,
+      "grad_norm": 2.8753548663225144,
+      "learning_rate": 9.538932721758474e-05,
+      "loss": 0.5992,
+      "num_input_tokens_seen": 6456328,
+      "step": 615
+    },
+    {
+      "epoch": 0.15967035797064125,
+      "grad_norm": 2.4013005755622467,
+      "learning_rate": 9.528680281157999e-05,
+      "loss": 0.587,
+      "num_input_tokens_seen": 6509024,
+      "step": 620
+    },
+    {
+      "epoch": 0.16095802214782384,
+      "grad_norm": 3.860358696946306,
+      "learning_rate": 9.518320736600943e-05,
+      "loss": 0.5836,
+      "num_input_tokens_seen": 6561336,
+      "step": 625
+    },
+    {
+      "epoch": 0.16224568632500644,
+      "grad_norm": 3.187917212328382,
+      "learning_rate": 9.507854333092063e-05,
+      "loss": 0.5913,
+      "num_input_tokens_seen": 6614024,
+      "step": 630
+    },
+    {
+      "epoch": 0.16353335050218903,
+      "grad_norm": 3.5342177024321586,
+      "learning_rate": 9.497281318163346e-05,
+      "loss": 0.5693,
+      "num_input_tokens_seen": 6666416,
+      "step": 635
+    },
+    {
+      "epoch": 0.16482101467937163,
+      "grad_norm": 3.90374612709263,
+      "learning_rate": 9.486601941868154e-05,
+      "loss": 0.572,
+      "num_input_tokens_seen": 6718200,
+      "step": 640
+    },
+    {
+      "epoch": 0.1661086788565542,
+      "grad_norm": 4.4270591027201665,
+      "learning_rate": 9.475816456775313e-05,
+      "loss": 0.6111,
+      "num_input_tokens_seen": 6771256,
+      "step": 645
+    },
+    {
+      "epoch": 0.1673963430337368,
+      "grad_norm": 5.04761388655614,
+      "learning_rate": 9.464925117963133e-05,
+      "loss": 0.5959,
+      "num_input_tokens_seen": 6824008,
+      "step": 650
+    },
+    {
+      "epoch": 0.1673963430337368,
+      "eval_loss": 0.5542036890983582,
+      "eval_runtime": 68.9048,
+      "eval_samples_per_second": 1.742,
+      "eval_steps_per_second": 0.435,
+      "num_input_tokens_seen": 6824008,
+      "step": 650
     }
   ],
   "logging_steps": 5,
   "max_steps": 3400,
-  "num_input_tokens_seen": 6299656,
+  "num_input_tokens_seen": 6824008,
   "num_train_epochs": 1,
   "save_steps": 50,
   "stateful_callbacks": {
@@ -1094,7 +1183,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 415517873799168.0,
+  "total_flos": 450087645085696.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:250d976171954308b1426496d02274452f4f7036ecb43184794d9d3e92b560db
-size 7352
+oid sha256:9b2ceb98c0b4a50d3909c4f866386ff0e5093b24fd71a2054110090af7b4ef0e
+size 7416