Model save

README.md

@@ -2,13 +2,9 @@
 license: other
 library_name: peft
 tags:
-- alignment-handbook
-- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
-datasets:
-- ruozhiba
 base_model: 01-ai/Yi-6B
 model-index:
 - name: Yi-6B-ruozhiba-5e-5-50
@@ -20,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Yi-6B-ruozhiba-5e-5-50
 
-This model is a fine-tuned version of [01-ai/Yi-6B](https://huggingface.co/01-ai/Yi-6B) on the ruozhiba dataset.
+This model is a fine-tuned version of [01-ai/Yi-6B](https://huggingface.co/01-ai/Yi-6B) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 3.0726
+- Loss: 3.0795
 
 ## Model description
 
@@ -58,10 +54,11 @@ The following hyperparameters were used during training:
 | 1.7645        | 3.0   | 165  | 1.9680          |
 | 1.4795        | 4.0   | 220  | 2.0760          |
 | 1.2467        | 5.0   | 275  | 2.2715          |
-| 0.7614        | 7.0   | 385  | 2.7865          |
-| 0.6237        | 8.0   | 440  | 2.9589          |
-| 0.5711        | 9.0   | 495  | 3.0581          |
-| 0.5884        | 10.0  | 550  | 3.0726          |
+| 1.0034        | 6.0   | 330  | 2.5656          |
+| 0.8124        | 7.0   | 385  | 2.8052          |
+| 0.6269        | 8.0   | 440  | 2.9866          |
+| 0.5743        | 9.0   | 495  | 3.0649          |
+| 0.5465        | 10.0  | 550  | 3.0795          |
 
 
 ### Framework versions

adapter_config.json

@@ -19,13 +19,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "down_proj",
-    "o_proj",
     "v_proj",
-    "k_proj",
+    "up_proj",
     "gate_proj",
-    "up_proj"
+    "o_proj",
+    "k_proj",
+    "q_proj",
+    "down_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9ce552cc5fb48a9312f62598453e6872694589dbd7643cd7f32eba154c51f85
+oid sha256:23be8b0ddd57a5e83512270ba5454048325043400bbbc75342fe824cfd1429de
 size 145287696

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 10.0,
-    "eval_loss": 3.072557210922241,
-    "eval_runtime": 4.9184,
+    "eval_loss": 3.079519748687744,
+    "eval_runtime": 4.9055,
     "eval_samples": 23,
-    "eval_samples_per_second": 4.676,
-    "eval_steps_per_second": 1.22,
-    "train_loss": 0.2617128190127286,
-    "train_runtime": 10832.3445,
+    "eval_samples_per_second": 4.689,
+    "eval_steps_per_second": 1.223,
+    "train_loss": 1.0206408483331855,
+    "train_runtime": 23074.2586,
     "train_samples": 217,
-    "train_samples_per_second": 0.2,
-    "train_steps_per_second": 0.051
+    "train_samples_per_second": 0.094,
+    "train_steps_per_second": 0.024
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "eval_loss": 3.072557210922241,
-    "eval_runtime": 4.9184,
+    "eval_loss": 3.079519748687744,
+    "eval_runtime": 4.9055,
     "eval_samples": 23,
-    "eval_samples_per_second": 4.676,
-    "eval_steps_per_second": 1.22
+    "eval_samples_per_second": 4.689,
+    "eval_steps_per_second": 1.223
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "train_loss": 0.2617128190127286,
-    "train_runtime": 10832.3445,
+    "train_loss": 1.0206408483331855,
+    "train_runtime": 23074.2586,
     "train_samples": 217,
-    "train_samples_per_second": 0.2,
-    "train_steps_per_second": 0.051
+    "train_samples_per_second": 0.094,
+    "train_steps_per_second": 0.024
 }

trainer_state.json

@@ -562,400 +562,414 @@
       "loss": 1.0034,
       "step": 328
     },
+    {
+      "epoch": 6.0,
+      "pls_score": 59.6,
+      "std": 4.059162475191158,
+      "step": 330
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.565575361251831,
+      "eval_runtime": 4.9106,
+      "eval_samples_per_second": 4.684,
+      "eval_steps_per_second": 1.222,
+      "step": 330
+    },
     {
       "epoch": 6.04,
       "learning_rate": 2.034664207670925e-05,
-      "loss": 0.8258,
+      "loss": 0.8958,
       "step": 332
     },
     {
       "epoch": 6.11,
       "learning_rate": 1.972463450245226e-05,
-      "loss": 0.8152,
+      "loss": 0.8214,
       "step": 336
     },
     {
       "epoch": 6.18,
       "learning_rate": 1.9106026612264316e-05,
-      "loss": 0.7462,
+      "loss": 0.7499,
       "step": 340
     },
     {
       "epoch": 6.25,
       "learning_rate": 1.84912170650422e-05,
-      "loss": 0.7995,
+      "loss": 0.7933,
       "step": 344
     },
     {
       "epoch": 6.33,
       "learning_rate": 1.7880602071858692e-05,
-      "loss": 0.7714,
+      "loss": 0.8006,
       "step": 348
     },
     {
       "epoch": 6.4,
       "learning_rate": 1.7274575140626318e-05,
-      "loss": 0.8581,
+      "loss": 0.7292,
       "step": 352
     },
     {
       "epoch": 6.47,
       "learning_rate": 1.667352682250298e-05,
-      "loss": 0.7631,
+      "loss": 0.8303,
       "step": 356
     },
     {
       "epoch": 6.55,
       "learning_rate": 1.6077844460203206e-05,
-      "loss": 0.8117,
+      "loss": 0.7661,
       "step": 360
     },
     {
       "epoch": 6.62,
       "learning_rate": 1.5487911938376924e-05,
-      "loss": 0.7725,
+      "loss": 0.8356,
       "step": 364
     },
     {
       "epoch": 6.69,
       "learning_rate": 1.4904109436216884e-05,
-      "loss": 0.8519,
+      "loss": 0.7956,
       "step": 368
     },
     {
       "epoch": 6.76,
       "learning_rate": 1.4326813182453958e-05,
-      "loss": 0.8037,
+      "loss": 0.7794,
       "step": 372
     },
     {
       "epoch": 6.84,
       "learning_rate": 1.3756395212898359e-05,
-      "loss": 0.848,
+      "loss": 0.7468,
       "step": 376
     },
     {
       "epoch": 6.91,
       "learning_rate": 1.3193223130682936e-05,
-      "loss": 0.797,
+      "loss": 0.8426,
       "step": 380
     },
     {
       "epoch": 6.98,
       "learning_rate": 1.2637659869363083e-05,
-      "loss": 0.7614,
+      "loss": 0.8124,
       "step": 384
     },
     {
       "epoch": 7.0,
-      "pls_score": 57.2,
-      "std": 4.3085032203771165,
+      "pls_score": 56.326530612244895,
+      "std": 4.027124583051197,
       "step": 385
     },
     {
       "epoch": 7.0,
-      "eval_loss": 2.7865102291107178,
-      "eval_runtime": 4.9204,
-      "eval_samples_per_second": 4.674,
-      "eval_steps_per_second": 1.219,
+      "eval_loss": 2.80517578125,
+      "eval_runtime": 4.9124,
+      "eval_samples_per_second": 4.682,
+      "eval_steps_per_second": 1.221,
       "step": 385
     },
     {
       "epoch": 7.05,
       "learning_rate": 1.2090063459025955e-05,
-      "loss": 0.8261,
+      "loss": 0.5562,
       "step": 388
     },
     {
       "epoch": 7.13,
       "learning_rate": 1.155078679555969e-05,
-      "loss": 0.639,
+      "loss": 0.6813,
       "step": 392
     },
     {
       "epoch": 7.2,
       "learning_rate": 1.1020177413231334e-05,
-      "loss": 0.748,
+      "loss": 0.6176,
       "step": 396
     },
     {
       "epoch": 7.27,
       "learning_rate": 1.049857726072005e-05,
-      "loss": 0.6049,
+      "loss": 0.5879,
       "step": 400
     },
     {
       "epoch": 7.35,
       "learning_rate": 9.986322480749927e-06,
-      "loss": 0.5971,
+      "loss": 0.6449,
       "step": 404
     },
     {
       "epoch": 7.42,
       "learning_rate": 9.483743193464408e-06,
-      "loss": 0.6648,
+      "loss": 0.6934,
       "step": 408
     },
     {
       "epoch": 7.49,
       "learning_rate": 8.991163283681944e-06,
-      "loss": 0.6513,
+      "loss": 0.7495,
       "step": 412
     },
     {
       "epoch": 7.56,
       "learning_rate": 8.508900192169964e-06,
-      "loss": 0.5917,
+      "loss": 0.6499,
       "step": 416
     },
     {
       "epoch": 7.64,
       "learning_rate": 8.0372647110717e-06,
-      "loss": 0.6839,
+      "loss": 0.6279,
       "step": 420
     },
     {
       "epoch": 7.71,
       "learning_rate": 7.576560783617668e-06,
-      "loss": 0.6658,
+      "loss": 0.6939,
       "step": 424
     },
     {
       "epoch": 7.78,
       "learning_rate": 7.127085308250914e-06,
-      "loss": 0.691,
+      "loss": 0.6682,
       "step": 428
     },
     {
       "epoch": 7.85,
       "learning_rate": 6.689127947292231e-06,
-      "loss": 0.7198,
+      "loss": 0.7422,
       "step": 432
     },
     {
       "epoch": 7.93,
       "learning_rate": 6.2629709402686535e-06,
-      "loss": 0.6533,
+      "loss": 0.6291,
       "step": 436
     },
     {
       "epoch": 8.0,
       "learning_rate": 5.848888922025553e-06,
-      "loss": 0.6237,
+      "loss": 0.6269,
       "step": 440
     },
     {
       "epoch": 8.0,
-      "pls_score": 60.4,
-      "std": 4.507416111254875,
+      "pls_score": 57.95918367346939,
+      "std": 4.290372510588257,
       "step": 440
     },
     {
       "epoch": 8.0,
-      "eval_loss": 2.958909511566162,
-      "eval_runtime": 4.9181,
-      "eval_samples_per_second": 4.677,
-      "eval_steps_per_second": 1.22,
+      "eval_loss": 2.9865965843200684,
+      "eval_runtime": 4.9092,
+      "eval_samples_per_second": 4.685,
+      "eval_steps_per_second": 1.222,
       "step": 440
     },
     {
       "epoch": 8.07,
       "learning_rate": 5.4471487457395225e-06,
-      "loss": 0.5912,
+      "loss": 0.6428,
       "step": 444
     },
     {
       "epoch": 8.15,
       "learning_rate": 5.058009310946119e-06,
-      "loss": 0.6374,
+      "loss": 0.6155,
       "step": 448
     },
     {
       "epoch": 8.22,
       "learning_rate": 4.681721396693303e-06,
-      "loss": 0.5975,
+      "loss": 0.5257,
       "step": 452
     },
     {
       "epoch": 8.29,
       "learning_rate": 4.318527499928074e-06,
-      "loss": 0.5448,
+      "loss": 0.5577,
       "step": 456
     },
     {
       "epoch": 8.36,
       "learning_rate": 3.968661679220468e-06,
-      "loss": 0.5574,
+      "loss": 0.6239,
       "step": 460
     },
     {
       "epoch": 8.44,
       "learning_rate": 3.632349403925664e-06,
-      "loss": 0.6206,
+      "loss": 0.6014,
       "step": 464
     },
     {
       "epoch": 8.51,
       "learning_rate": 3.3098074088812686e-06,
-      "loss": 0.6207,
+      "loss": 0.5279,
       "step": 468
     },
     {
       "epoch": 8.58,
       "learning_rate": 3.0012435547336737e-06,
-      "loss": 0.5411,
+      "loss": 0.5482,
       "step": 472
     },
     {
       "epoch": 8.65,
       "learning_rate": 2.7068566939831645e-06,
-      "loss": 0.6231,
+      "loss": 0.5768,
       "step": 476
     },
     {
       "epoch": 8.73,
       "learning_rate": 2.4268365428344736e-06,
-      "loss": 0.5944,
+      "loss": 0.5444,
       "step": 480
     },
     {
       "epoch": 8.8,
       "learning_rate": 2.1613635589349756e-06,
-      "loss": 0.6279,
+      "loss": 0.674,
       "step": 484
     },
     {
       "epoch": 8.87,
       "learning_rate": 1.9106088250797267e-06,
-      "loss": 0.561,
+      "loss": 0.6033,
       "step": 488
     },
     {
       "epoch": 8.95,
       "learning_rate": 1.674733938957873e-06,
-      "loss": 0.5711,
+      "loss": 0.5743,
       "step": 492
     },
     {
       "epoch": 9.0,
-      "pls_score": 55.2,
-      "std": 4.365684367885522,
+      "pls_score": 55.51020408163265,
+      "std": 4.29116489632049,
       "step": 495
     },
     {
       "epoch": 9.0,
-      "eval_loss": 3.058058977127075,
-      "eval_runtime": 4.9251,
-      "eval_samples_per_second": 4.67,
-      "eval_steps_per_second": 1.218,
+      "eval_loss": 3.064878463745117,
+      "eval_runtime": 4.9105,
+      "eval_samples_per_second": 4.684,
+      "eval_steps_per_second": 1.222,
       "step": 495
     },
     {
       "epoch": 9.02,
       "learning_rate": 1.4538909090118846e-06,
-      "loss": 0.5386,
+      "loss": 0.5749,
       "step": 496
     },
     {
       "epoch": 9.09,
       "learning_rate": 1.248222056476367e-06,
-      "loss": 0.6031,
+      "loss": 0.6153,
       "step": 500
     },
     {
       "epoch": 9.16,
       "learning_rate": 1.0578599236598707e-06,
-      "loss": 0.5676,
+      "loss": 0.5336,
       "step": 504
     },
     {
       "epoch": 9.24,
       "learning_rate": 8.829271885286094e-07,
-      "loss": 0.5298,
+      "loss": 0.519,
       "step": 508
     },
     {
       "epoch": 9.31,
       "learning_rate": 7.235365856472442e-07,
-      "loss": 0.5428,
+      "loss": 0.6333,
       "step": 512
     },
     {
       "epoch": 9.38,
       "learning_rate": 5.797908335276214e-07,
-      "loss": 0.5811,
+      "loss": 0.5356,
       "step": 516
     },
     {
       "epoch": 9.45,
       "learning_rate": 4.517825684323324e-07,
-      "loss": 0.5703,
+      "loss": 0.557,
       "step": 520
     },
     {
       "epoch": 9.53,
       "learning_rate": 3.395942846757066e-07,
-      "loss": 0.6147,
+      "loss": 0.6558,
       "step": 524
     },
     {
       "epoch": 9.6,
       "learning_rate": 2.4329828146074095e-07,
-      "loss": 0.5451,
+      "loss": 0.5504,
       "step": 528
     },
     {
       "epoch": 9.67,
       "learning_rate": 1.6295661628624447e-07,
-      "loss": 0.5303,
+      "loss": 0.5123,
       "step": 532
     },
     {
       "epoch": 9.75,
       "learning_rate": 9.862106495415469e-08,
-      "loss": 0.5629,
+      "loss": 0.5928,
       "step": 536
     },
     {
       "epoch": 9.82,
       "learning_rate": 5.033308820289184e-08,
-      "loss": 0.5472,
+      "loss": 0.5185,
       "step": 540
     },
     {
       "epoch": 9.89,
       "learning_rate": 1.812380498815991e-08,
-      "loss": 0.5334,
+      "loss": 0.5582,
       "step": 544
     },
     {
       "epoch": 9.96,
       "learning_rate": 2.0139724285161977e-09,
-      "loss": 0.5884,
+      "loss": 0.5465,
       "step": 548
     },
     {
       "epoch": 10.0,
-      "pls_score": 60.0,
-      "std": 4.195235392680606,
+      "pls_score": 53.06122448979592,
+      "std": 4.340793602433762,
       "step": 550
     },
     {
       "epoch": 10.0,
-      "eval_loss": 3.072557210922241,
-      "eval_runtime": 4.9196,
-      "eval_samples_per_second": 4.675,
-      "eval_steps_per_second": 1.22,
+      "eval_loss": 3.079519748687744,
+      "eval_runtime": 4.9095,
+      "eval_samples_per_second": 4.685,
+      "eval_steps_per_second": 1.222,
       "step": 550
     },
     {
       "epoch": 10.0,
       "step": 550,
-      "total_flos": 1.870637420231885e+16,
-      "train_loss": 0.2617128190127286,
-      "train_runtime": 10832.3445,
-      "train_samples_per_second": 0.2,
-      "train_steps_per_second": 0.051
+      "total_flos": 1.876652342808576e+16,
+      "train_loss": 1.0206408483331855,
+      "train_runtime": 23074.2586,
+      "train_samples_per_second": 0.094,
+      "train_steps_per_second": 0.024
     }
   ],
   "logging_steps": 4,
@@ -963,7 +977,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 55,
-  "total_flos": 1.870637420231885e+16,
+  "total_flos": 1.876652342808576e+16,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0bc819b0296221e62114b2956d623b867b1651f22681fcdf7256a558db58908a
+oid sha256:7f9b09f24f51fddb4e163c6d8b5dda5a40a8a6543c084d147043f7abd98091e1
 size 4792