Training in progress, step 4, checkpoint

last-checkpoint/adapter_config.json

@@ -10,22 +10,22 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 16,
-  "lora_dropout": 0.1,
+  "lora_alpha": 32,
+  "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": [
-    "k_proj",
+    "gate_proj",
     "up_proj",
-    "v_proj",
     "down_proj",
     "q_proj",
-    "gate_proj",
+    "v_proj",
+    "k_proj",
     "o_proj"
   ],
   "task_type": "CAUSAL_LM",

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f09978d4032be96d17ac53467ffdca8f81c03e97f7f717957ca15ff29f09c81a
-size 83945296
+oid sha256:5e0d4550640c26f65dc3d4095aaf17861a63a272705a71fa441198798a5f5f9c
+size 167832240

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:967c568d890ad6e03543b1d3f26f368322e126bc130bbdd9b9a43e0f169d5aa5
-size 168149074
+oid sha256:1657c1fbb95227227e5bb764f2408b5c35d1c1cb6bfdade7499f2ec873e276b2
+size 335922386

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:05531349f6e2487143a7251afc8a645033108171aa1d481234785012b46d00ef
+oid sha256:b274d200228065c5ad050511a1f7866a6edb57bdaa953138cf781ba60f5569eb
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ffd8c58e5d02492554dbaa495f8cf80dff41fabc0e1288cb2fd18a103d402219
+oid sha256:917587e2c0a52fd7c3d6680e02e388879c84c249a215f319208d6a9951b36ba9
 size 1064

last-checkpoint/trainer_state.json

@@ -1,16 +1,16 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.00027200398939184443,
-  "eval_steps": 3,
-  "global_step": 3,
+  "epoch": 0.0003626719858557926,
+  "eval_steps": 4,
+  "global_step": 4,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 9.066799646394814e-05,
-      "grad_norm": 0.9196107983589172,
+      "grad_norm": 1.305655837059021,
       "learning_rate": 2e-05,
       "loss": 0.9466,
       "step": 1
@@ -18,39 +18,46 @@
     {
       "epoch": 9.066799646394814e-05,
       "eval_loss": 1.1514371633529663,
-      "eval_runtime": 280.9406,
-      "eval_samples_per_second": 8.265,
-      "eval_steps_per_second": 8.265,
+      "eval_runtime": 281.757,
+      "eval_samples_per_second": 8.241,
+      "eval_steps_per_second": 8.241,
       "step": 1
     },
     {
       "epoch": 0.0001813359929278963,
-      "grad_norm": 1.2303106784820557,
+      "grad_norm": 1.7540494203567505,
       "learning_rate": 4e-05,
       "loss": 1.2689,
       "step": 2
     },
     {
       "epoch": 0.00027200398939184443,
-      "grad_norm": 0.953157901763916,
+      "grad_norm": 1.3806679248809814,
       "learning_rate": 6e-05,
-      "loss": 0.9876,
+      "loss": 0.9866,
       "step": 3
     },
     {
-      "epoch": 0.00027200398939184443,
-      "eval_loss": 1.1419087648391724,
-      "eval_runtime": 281.2312,
-      "eval_samples_per_second": 8.257,
-      "eval_steps_per_second": 8.257,
-      "step": 3
+      "epoch": 0.0003626719858557926,
+      "grad_norm": 1.6540546417236328,
+      "learning_rate": 8e-05,
+      "loss": 0.886,
+      "step": 4
+    },
+    {
+      "epoch": 0.0003626719858557926,
+      "eval_loss": 1.063585877418518,
+      "eval_runtime": 281.615,
+      "eval_samples_per_second": 8.245,
+      "eval_steps_per_second": 8.245,
+      "step": 4
     }
   ],
   "logging_steps": 1,
-  "max_steps": 25,
+  "max_steps": 50,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
-  "save_steps": 3,
+  "save_steps": 4,
   "stateful_callbacks": {
     "TrainerControl": {
       "args": {
@@ -63,7 +70,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 1109757876240384.0,
+  "total_flos": 1483800336924672.0,
   "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:bf5f0cf90aa3deddbb881f441c0237bff1fd72ac4c2cbf0e67c55438798ba345
+oid sha256:586972196f15dd32a3d87fb4e0aad2c5279b9a281041e5bf0b54d207daa83d8a
 size 6776