Training in progress, step 3, checkpoint

last-checkpoint/adapter_config.json

@@ -10,23 +10,23 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 8,
-  "lora_dropout": 0.1,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 4,
+  "r": 8,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "down_proj",
-    "up_proj",
     "v_proj",
-    "k_proj",
-    "gate_proj",
     "o_proj",
-    "q_proj"
+    "gate_proj",
+    "q_proj",
+    "up_proj",
+    "down_proj",
+    "k_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7e711b7b4034b2f3bd9799e3738e807081a18d9dd9d4402e426cefbb04d49c9
-size 9048
+oid sha256:ab90251ab6e27a57801a12d18af11f2f9f07fcab223af845d95de1a0b405501c
+size 14696

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e58e24af97e29d770d88764fcd1061468363d04933ab5bc770d00cb8fc8b69f1
-size 28134
+oid sha256:7b12248ecbc96221e2afdb2d474ebe9b42802236a73a19ebb91f18881b9c511f
+size 39398

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b789b44e35b8f0004dd2e31bfb1af056b8528eeb7ffa2f97d314e99f11187ccd
+oid sha256:464e6cdff3f8548ba79ace3859c72e6b06d1e045e69932eb75c584ad0e8918c8
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6da16eefab22daee3fb0999c7651be192feb36b73810bf419e9165c7023ec02
+oid sha256:ffd8c58e5d02492554dbaa495f8cf80dff41fabc0e1288cb2fd18a103d402219
 size 1064

last-checkpoint/trainer_state.json

@@ -1,34 +1,56 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.0002666666666666667,
-  "eval_steps": 500,
-  "global_step": 1,
+  "epoch": 0.0016825574873808188,
+  "eval_steps": 3,
+  "global_step": 3,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.0002666666666666667,
-      "grad_norm": 0.014549034647643566,
-      "learning_rate": 0.0,
-      "loss": 11.9308,
+      "epoch": 0.0005608524957936063,
+      "grad_norm": 0.014698871411383152,
+      "learning_rate": 2e-05,
+      "loss": 11.9315,
       "step": 1
     },
     {
-      "epoch": 0.0002666666666666667,
-      "eval_loss": 11.933904647827148,
-      "eval_runtime": 0.0581,
-      "eval_samples_per_second": 275.597,
-      "eval_steps_per_second": 68.899,
+      "epoch": 0.0005608524957936063,
+      "eval_loss": 11.931306838989258,
+      "eval_runtime": 2.6652,
+      "eval_samples_per_second": 281.781,
+      "eval_steps_per_second": 141.078,
       "step": 1
+    },
+    {
+      "epoch": 0.0011217049915872126,
+      "grad_norm": 0.009716438129544258,
+      "learning_rate": 4e-05,
+      "loss": 11.9304,
+      "step": 2
+    },
+    {
+      "epoch": 0.0016825574873808188,
+      "grad_norm": 0.013599561527371407,
+      "learning_rate": 6e-05,
+      "loss": 11.9319,
+      "step": 3
+    },
+    {
+      "epoch": 0.0016825574873808188,
+      "eval_loss": 11.931304931640625,
+      "eval_runtime": 2.7015,
+      "eval_samples_per_second": 277.993,
+      "eval_steps_per_second": 139.182,
+      "step": 3
     }
   ],
   "logging_steps": 1,
-  "max_steps": 1,
+  "max_steps": 10,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
-  "save_steps": 500,
+  "save_steps": 3,
   "stateful_callbacks": {
     "TrainerControl": {
       "args": {
@@ -36,13 +58,13 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": true
+        "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 10444800.0,
-  "train_batch_size": 4,
+  "total_flos": 384024576.0,
+  "train_batch_size": 2,
   "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:a34c87415c9e639533cb4d7b940fdf4014edaec46abd1ff3438eb94417519db3
+oid sha256:525897d6fc230472c14ab0a00d3e93c0465a5d3840498e66ad09e713ff0af362
 size 6776