Training in progress, step 8, 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 sha256:8050ea3e8bd38ef53a91f964634cca0140e5bca7df70f46096159c5b9a35982c
-size 83945296
+oid sha256:d7db16dafee18a681cd947e435ccf1bc80930f655be31343b8c30d91425c5298
+size 167832240

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5a690058fbd348365b2bb346924b7dec8274ba59671ba9df584e9971ae9f33b9
-size 168149074
+oid sha256:e8d429c9cc6d650d3f05af7f42c23e8a30f7115fbaf81ec5d139884095fd1904
+size 335922386

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce79b427bf34068a1374bc4838a962d39a0a665c827f62ca40d1c2d968d87cf3
+oid sha256:d64da661c36952fedef4f75a5b26e1dead99ce0e439409ef2a0ead860962e6ec
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0099cb7287625b29b67c4fcf42ff20fae623b429bfb10f5ac695bc54f2be54fd
+oid sha256:26a93b2a1f4b5368650119fe6e0d6eec6d19cda6badeba4d21943ab48964fa00
 size 1064

last-checkpoint/trainer_state.json

@@ -1,16 +1,16 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.0005440079787836889,
-  "eval_steps": 3,
-  "global_step": 6,
+  "epoch": 0.0007253439717115851,
+  "eval_steps": 4,
+  "global_step": 8,
   "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
@@ -18,68 +18,82 @@
     {
       "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,
-      "grad_norm": 1.2283321619033813,
-      "learning_rate": 8e-05,
-      "loss": 0.8963,
+      "eval_loss": 1.063585877418518,
+      "eval_runtime": 281.615,
+      "eval_samples_per_second": 8.245,
+      "eval_steps_per_second": 8.245,
       "step": 4
     },
     {
       "epoch": 0.00045333998231974066,
-      "grad_norm": 1.0397660732269287,
+      "grad_norm": 1.351820945739746,
       "learning_rate": 0.0001,
-      "loss": 1.064,
+      "loss": 1.0148,
       "step": 5
     },
     {
       "epoch": 0.0005440079787836889,
-      "grad_norm": 1.3788177967071533,
+      "grad_norm": 1.6633617877960205,
       "learning_rate": 0.00012,
-      "loss": 1.311,
+      "loss": 1.1922,
       "step": 6
     },
     {
-      "epoch": 0.0005440079787836889,
-      "eval_loss": 0.9817929267883301,
-      "eval_runtime": 281.7359,
-      "eval_samples_per_second": 8.242,
-      "eval_steps_per_second": 8.242,
-      "step": 6
+      "epoch": 0.000634675975247637,
+      "grad_norm": 1.216033697128296,
+      "learning_rate": 0.00014,
+      "loss": 0.7838,
+      "step": 7
+    },
+    {
+      "epoch": 0.0007253439717115851,
+      "grad_norm": 1.0886664390563965,
+      "learning_rate": 0.00016,
+      "loss": 0.6905,
+      "step": 8
+    },
+    {
+      "epoch": 0.0007253439717115851,
+      "eval_loss": 0.6156781911849976,
+      "eval_runtime": 282.2824,
+      "eval_samples_per_second": 8.226,
+      "eval_steps_per_second": 8.226,
+      "step": 8
     }
   ],
   "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": {
@@ -92,7 +106,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2311995575500800.0,
+  "total_flos": 3060338194907136.0,
   "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