Training in progress, step 3401, checkpoint

last-checkpoint/adapter_config.json

@@ -19,7 +19,7 @@
   "r": 8,
   "rank_pattern": {},
   "revision": null,
-  "target_modules": "^(?!.*patch_embed).*(?:q_proj|qkv|gate_proj|fc1|up_proj|k_proj|o_proj|fc2|down_proj|v_proj|proj).*",
+  "target_modules": "^(?!.*patch_embed).*(?:proj|v_proj|fc1|gate_proj|down_proj|o_proj|fc2|k_proj|q_proj|up_proj|qkv).*",
   "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:d45234d4f1bea3302aef4c530a94a48449dbe78e014c5812413b6eb5ae04f226
+oid sha256:302b9ce048dad99dc59725f9bc543136929b03fd0548bef61608ff98a26b885e
 size 29034840

last-checkpoint/global_step3401/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:860876f98bf03c575d9bc5921783fd219c7ce0e0f47f53cc31e943c2bc5f81dc
+size 43429616

last-checkpoint/global_step3401/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:af76082060d99cabb6cf54a5de33e91f9afcbe1cb03f27effeb57c5272eeba35
+size 43429616

last-checkpoint/global_step3401/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:c2ebe21d90d4da9ac7c8938d709c4e4ac5a423f047f062156c5cc024ec022684
+size 43429616

last-checkpoint/global_step3401/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a9dd47364015c06839e58d41804e93f4f3eb84000c395948f049763066d044c
+size 43429616

last-checkpoint/global_step3401/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70455332cd654c8606fdab3c958e34cfb2dfc91b03c03e361d5d2e7734630486
+size 637299

last-checkpoint/global_step3401/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:570bfba1d3e07de3836b00bdea62b95fe6d5d0e0e65c10bd9766a23c4f6ac8c5
+size 637171

last-checkpoint/global_step3401/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:984640bbf0cdbe7ead13f1a20c28e262fcc60b3d15d4abc95f25a927d576c7ca
+size 637171

last-checkpoint/global_step3401/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:305bc562bdc0a7614ebce33e5124dbeba838eea4588e98e9e12061348423eec5
+size 637171

last-checkpoint/latest

@@ -1 +1 @@
-global_step3400
+global_step3401

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:125e4e6f7ec60dee414409e1dcb067d92d3360952c80aae0fd0e141af3db4b0d
+oid sha256:3f2d6472ebcb070058b4055bf37ad09560e10692efc14730a79abd74cca6a737
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": 0.1869634985923767,
   "best_model_checkpoint": "saves/CADICA_qwenvl_direction_then_DetectAndClassify_scale6/lora/sft/checkpoint-2350",
-  "epoch": 1.0044313146233383,
+  "epoch": 1.0047267355982274,
   "eval_steps": 50,
-  "global_step": 3400,
+  "global_step": 3401,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -6062,8 +6062,8 @@
     }
   ],
   "logging_steps": 5,
-  "max_steps": 6770,
-  "num_input_tokens_seen": 35305984,
+  "max_steps": 3400,
+  "num_input_tokens_seen": 35316128,
   "num_train_epochs": 2,
   "save_steps": 50,
   "stateful_callbacks": {
@@ -6073,12 +6073,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
-  "total_flos": 2329291457560576.0,
+  "total_flos": 2329910849044480.0,
   "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 sha256:157f40a64e5df8b268b803366ba3fce3e3b5e6e2f20d7c81bfc44726f0bee4e8
+oid sha256:85339a802dfd06cbba7d2568099cd0116c5715fbc57a8d66871ebfbd39c462c6
 size 7480