Training in progress, epoch 0, checkpoint

checkpoint-10/adapter_config.json

@@ -9,13 +9,13 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 8,
+  "lora_alpha": 16,
   "lora_dropout": 0.05,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 8,
+  "r": 16,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [

checkpoint-10/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9b2d576eceb9eb6d2d084808671f66357fe38db8508ae6def2045d2e86c4681
-size 13648432
+oid sha256:41b0915dbabe732a22a164ddd732f0bc39f270aabadd91d0af928396b0fd8cda
+size 27280152

checkpoint-10/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4683358d3ecdb4f1484131e255d12b2c508a88328d6b29e71db7aeee7c48619f
-size 27338682
+oid sha256:fef419d6c9ebc6eff293f10d20b8d172f6c661b12092e1e0f2c11c32a2cad3be
+size 54602042

checkpoint-10/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd0704cf53e73fdd3790d041fcd30775fdd963463e54c18c6c10f750649e8839
+oid sha256:1e119516d1e041105639f70ebe026fb0f629d45578e5ab3a60aa8f90c988f825
 size 14244

checkpoint-10/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.5767254829406738,
+  "best_metric": 0.5605378746986389,
   "best_model_checkpoint": "./Zephyr/12-01-24-HuggingFaceH4-zephyr-7b-beta_Treino-GPTQ-2_epochs-5_batch_1/checkpoints/checkpoint-10",
-  "epoch": 0.06938421509106678,
+  "epoch": 0.2775368603642671,
   "eval_steps": 500,
   "global_step": 10,
   "is_hyper_param_search": false,
@@ -9,11 +9,11 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.07,
-      "eval_loss": 0.5767254829406738,
-      "eval_runtime": 374.7385,
-      "eval_samples_per_second": 0.342,
-      "eval_steps_per_second": 0.171,
+      "epoch": 0.28,
+      "eval_loss": 0.5605378746986389,
+      "eval_runtime": 375.7067,
+      "eval_samples_per_second": 0.341,
+      "eval_steps_per_second": 0.17,
       "step": 10
     }
   ],
@@ -21,7 +21,7 @@
   "max_steps": 10,
   "num_train_epochs": 1,
   "save_steps": 500,
-  "total_flos": 2.797365350105088e+16,
+  "total_flos": 1.119482151960576e+17,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-10/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf8ad3c191012642f1f5613c45fec6745c7e550e2a431878ff850248c0965e9f
+oid sha256:bf37fa4263bd42407013461e43287124efa1cc33f6427e440772692788437888
 size 4856