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-ruozhiba3
@@ -20,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Yi-6B-ruozhiba3
 
-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: 4.3351
 
@@ -48,7 +44,7 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 3
+- num_epochs: 20
 
 ### Training results
 

adapter_config.json

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

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 20.0,
     "eval_loss": 4.335062026977539,
-    "eval_runtime": 6.4371,
+    "eval_runtime": 6.2462,
     "eval_samples": 23,
-    "eval_samples_per_second": 3.573,
-    "eval_steps_per_second": 0.932,
+    "eval_samples_per_second": 3.682,
+    "eval_steps_per_second": 0.961,
     "train_loss": 0.0,
-    "train_runtime": 10.2256,
+    "train_runtime": 10.1895,
     "train_samples": 217,
-    "train_samples_per_second": 63.664,
-    "train_steps_per_second": 16.136
+    "train_samples_per_second": 425.927,
+    "train_steps_per_second": 107.954
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 20.0,
     "eval_loss": 4.335062026977539,
-    "eval_runtime": 6.4371,
+    "eval_runtime": 6.2462,
     "eval_samples": 23,
-    "eval_samples_per_second": 3.573,
-    "eval_steps_per_second": 0.932
+    "eval_samples_per_second": 3.682,
+    "eval_steps_per_second": 0.961
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 20.0,
     "train_loss": 0.0,
-    "train_runtime": 10.2256,
+    "train_runtime": 10.1895,
     "train_samples": 217,
-    "train_samples_per_second": 63.664,
-    "train_steps_per_second": 16.136
+    "train_samples_per_second": 425.927,
+    "train_steps_per_second": 107.954
 }

trainer_state.json

@@ -501,15 +501,15 @@
       "step": 1100,
       "total_flos": 3.807078373542298e+16,
       "train_loss": 0.0,
-      "train_runtime": 10.2256,
-      "train_samples_per_second": 63.664,
-      "train_steps_per_second": 16.136
+      "train_runtime": 10.1895,
+      "train_samples_per_second": 425.927,
+      "train_steps_per_second": 107.954
     }
   ],
   "logging_steps": 20,
-  "max_steps": 165,
+  "max_steps": 1100,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 3,
+  "num_train_epochs": 20,
   "save_steps": 20,
   "total_flos": 3.807078373542298e+16,
   "train_batch_size": 4,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b591bed5d250490d702e6db47039b4083880a6eccfa19736ca1c3fe16762de21
+oid sha256:d9f736ab86e14c5ce1673e8cdbda8ef77a9397529a9bc1b33bab2a4b8bbb8729
 size 4728