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
@@ -20,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Yi-6B-ruozhiba
 
-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: 2.2082
+- Loss: 2.2134
 
 ## Model description
 
@@ -42,8 +38,8 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 1
-- eval_batch_size: 1
+- train_batch_size: 4
+- eval_batch_size: 4
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
@@ -58,7 +54,7 @@ The following hyperparameters were used during training:
 | 1.7781        | 2.0   | 434  | 1.9393          |
 | 1.4563        | 3.0   | 651  | 2.0187          |
 | 1.3206        | 4.0   | 868  | 2.1767          |
-| 1.1018        | 5.0   | 1085 | 2.2082          |
+| 1.1018        | 5.0   | 1085 | 2.2134          |
 
 
 ### Framework versions

adapter_config.json

@@ -19,12 +19,12 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "up_proj",
     "q_proj",
     "k_proj",
+    "v_proj",
+    "up_proj",
     "o_proj",
     "down_proj",
-    "v_proj",
     "gate_proj"
   ],
   "task_type": "CAUSAL_LM"

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ff29234f0bb30543464da6cc219166f299f4a42af925bef04b8a8aa2aa26d45c
+oid sha256:b6213550ad2eaebefc2518e7f4cce91f4a1c939228aefb0bb11a36d14883e131
 size 72673912

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 5.0,
-    "eval_loss": 2.2082066535949707,
-    "eval_runtime": 2.7423,
+    "eval_loss": 2.2133874893188477,
+    "eval_runtime": 1.2237,
     "eval_samples": 23,
-    "eval_samples_per_second": 8.387,
-    "eval_steps_per_second": 8.387,
-    "train_loss": 1.4582410645375055,
-    "train_runtime": 606.5736,
+    "eval_samples_per_second": 18.796,
+    "eval_steps_per_second": 4.903,
+    "train_loss": 0.005994410009428104,
+    "train_runtime": 16.2081,
     "train_samples": 217,
-    "train_samples_per_second": 1.789,
-    "train_steps_per_second": 1.789
+    "train_samples_per_second": 66.942,
+    "train_steps_per_second": 66.942
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
-    "eval_loss": 2.2082066535949707,
-    "eval_runtime": 2.7423,
+    "eval_loss": 2.2133874893188477,
+    "eval_runtime": 1.2237,
     "eval_samples": 23,
-    "eval_samples_per_second": 8.387,
-    "eval_steps_per_second": 8.387
+    "eval_samples_per_second": 18.796,
+    "eval_steps_per_second": 4.903
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
-    "train_loss": 1.4582410645375055,
-    "train_runtime": 606.5736,
+    "train_loss": 0.005994410009428104,
+    "train_runtime": 16.2081,
     "train_samples": 217,
-    "train_samples_per_second": 1.789,
-    "train_steps_per_second": 1.789
+    "train_samples_per_second": 66.942,
+    "train_steps_per_second": 66.942
 }

trainer_state.json

@@ -210,20 +210,20 @@
     },
     {
       "epoch": 5.0,
-      "eval_loss": 2.2082066535949707,
-      "eval_runtime": 2.7308,
-      "eval_samples_per_second": 8.422,
-      "eval_steps_per_second": 8.422,
+      "eval_loss": 2.2133874893188477,
+      "eval_runtime": 1.2338,
+      "eval_samples_per_second": 18.641,
+      "eval_steps_per_second": 4.863,
       "step": 1085
     },
     {
       "epoch": 5.0,
       "step": 1085,
-      "total_flos": 6971428308910080.0,
-      "train_loss": 1.4582410645375055,
-      "train_runtime": 606.5736,
-      "train_samples_per_second": 1.789,
-      "train_steps_per_second": 1.789
+      "total_flos": 6960819918102528.0,
+      "train_loss": 0.005994410009428104,
+      "train_runtime": 16.2081,
+      "train_samples_per_second": 66.942,
+      "train_steps_per_second": 66.942
     }
   ],
   "logging_steps": 40,
@@ -231,7 +231,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 5,
   "save_steps": 20,
-  "total_flos": 6971428308910080.0,
+  "total_flos": 6960819918102528.0,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd38b2b4154677730d8c95476097e2b283f6a5b731cedb5377ece99a75ca8c2f
+oid sha256:db0e82d241388aadcc3b6e4d4719f8a3ba21846f290c5cad89f680cd5fc579a4
 size 4728