Model save

README.md

@@ -3,15 +3,11 @@ library_name: transformers
 license: apache-2.0
 base_model: alignment-handbook/zephyr-7b-sft-full
 tags:
-- alignment-handbook
-- trl
-- sft
-- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- scottsuk0306/DepthQA
+- generator
 model-index:
 - name: zephyr-7b-sft-full
   results: []
@@ -22,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-sft-full
 
-This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the scottsuk0306/DepthQA dataset.
+This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the generator dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4854
+- Loss: 0.4115
 
 ## Model description
 
@@ -60,16 +56,16 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 1.1139        | 1.0   | 1    | 1.1133          |
-| 1.1139        | 2.0   | 2    | 1.2947          |
-| 1.1139        | 3.0   | 3    | 1.0839          |
-| 1.1139        | 4.0   | 4    | 0.8506          |
-| 1.0867        | 5.0   | 5    | 0.7521          |
-| 1.0867        | 6.0   | 6    | 0.6283          |
-| 1.0867        | 7.0   | 7    | 0.5677          |
-| 1.0867        | 8.0   | 8    | 0.5233          |
-| 1.0867        | 9.0   | 9    | 0.4948          |
-| 0.5927        | 10.0  | 10   | 0.4854          |
+| 1.1097        | 1.0   | 1    | 1.1095          |
+| 1.1097        | 2.0   | 2    | 1.2840          |
+| 1.1097        | 3.0   | 3    | 1.0815          |
+| 1.1097        | 4.0   | 4    | 0.8740          |
+| 1.0878        | 5.0   | 5    | 0.7459          |
+| 1.0878        | 6.0   | 6    | 0.6034          |
+| 1.0878        | 7.0   | 7    | 0.5202          |
+| 1.0878        | 8.0   | 8    | 0.4572          |
+| 1.0878        | 9.0   | 9    | 0.4221          |
+| 0.5493        | 10.0  | 10   | 0.4115          |
 
 
 ### Framework versions

all_results.json

@@ -1,14 +1,9 @@
 {
     "epoch": 10.0,
-    "eval_loss": 0.48538458347320557,
-    "eval_runtime": 2.6719,
-    "eval_samples": 851,
-    "eval_samples_per_second": 17.965,
-    "eval_steps_per_second": 0.374,
     "total_flos": 8375186227200.0,
-    "train_loss": 0.8424163699150086,
-    "train_runtime": 304.1051,
-    "train_samples": 851,
-    "train_samples_per_second": 1.578,
-    "train_steps_per_second": 0.033
+    "train_loss": 0.8207268595695496,
+    "train_runtime": 309.7037,
+    "train_samples": 848,
+    "train_samples_per_second": 1.614,
+    "train_steps_per_second": 0.032
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 10.0,
     "total_flos": 8375186227200.0,
-    "train_loss": 0.8424163699150086,
-    "train_runtime": 304.1051,
-    "train_samples": 851,
-    "train_samples_per_second": 1.578,
-    "train_steps_per_second": 0.033
+    "train_loss": 0.8207268595695496,
+    "train_runtime": 309.7037,
+    "train_samples": 848,
+    "train_samples_per_second": 1.614,
+    "train_steps_per_second": 0.032
 }

trainer_state.json

@@ -10,113 +10,113 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 18.364192038860114,
+      "grad_norm": 17.810282459441467,
       "learning_rate": 1e-05,
-      "loss": 1.1139,
+      "loss": 1.1097,
       "step": 1
     },
     {
       "epoch": 1.0,
-      "eval_loss": 1.1133455038070679,
-      "eval_runtime": 2.8092,
-      "eval_samples_per_second": 17.087,
-      "eval_steps_per_second": 0.356,
+      "eval_loss": 1.1095243692398071,
+      "eval_runtime": 2.8023,
+      "eval_samples_per_second": 17.843,
+      "eval_steps_per_second": 0.357,
       "step": 1
     },
     {
       "epoch": 2.0,
-      "eval_loss": 1.2946728467941284,
-      "eval_runtime": 2.6987,
-      "eval_samples_per_second": 17.786,
-      "eval_steps_per_second": 0.371,
+      "eval_loss": 1.2839607000350952,
+      "eval_runtime": 2.6692,
+      "eval_samples_per_second": 18.732,
+      "eval_steps_per_second": 0.375,
       "step": 2
     },
     {
       "epoch": 3.0,
-      "eval_loss": 1.0838984251022339,
-      "eval_runtime": 2.6695,
-      "eval_samples_per_second": 17.981,
-      "eval_steps_per_second": 0.375,
+      "eval_loss": 1.0814800262451172,
+      "eval_runtime": 2.6746,
+      "eval_samples_per_second": 18.695,
+      "eval_steps_per_second": 0.374,
       "step": 3
     },
     {
       "epoch": 4.0,
-      "eval_loss": 0.8506386280059814,
-      "eval_runtime": 2.6924,
-      "eval_samples_per_second": 17.828,
-      "eval_steps_per_second": 0.371,
+      "eval_loss": 0.874021053314209,
+      "eval_runtime": 2.6724,
+      "eval_samples_per_second": 18.71,
+      "eval_steps_per_second": 0.374,
       "step": 4
     },
     {
       "epoch": 5.0,
-      "grad_norm": 19.328671929154766,
+      "grad_norm": 18.20399999295951,
       "learning_rate": 5.8682408883346535e-06,
-      "loss": 1.0867,
+      "loss": 1.0878,
       "step": 5
     },
     {
       "epoch": 5.0,
-      "eval_loss": 0.7521028518676758,
-      "eval_runtime": 2.6708,
-      "eval_samples_per_second": 17.972,
-      "eval_steps_per_second": 0.374,
+      "eval_loss": 0.7459021210670471,
+      "eval_runtime": 2.6898,
+      "eval_samples_per_second": 18.589,
+      "eval_steps_per_second": 0.372,
       "step": 5
     },
     {
       "epoch": 6.0,
-      "eval_loss": 0.628281831741333,
-      "eval_runtime": 2.7091,
-      "eval_samples_per_second": 17.718,
-      "eval_steps_per_second": 0.369,
+      "eval_loss": 0.6033510565757751,
+      "eval_runtime": 2.6879,
+      "eval_samples_per_second": 18.602,
+      "eval_steps_per_second": 0.372,
       "step": 6
     },
     {
       "epoch": 7.0,
-      "eval_loss": 0.5677043795585632,
-      "eval_runtime": 2.6968,
-      "eval_samples_per_second": 17.799,
-      "eval_steps_per_second": 0.371,
+      "eval_loss": 0.5201848745346069,
+      "eval_runtime": 2.6772,
+      "eval_samples_per_second": 18.676,
+      "eval_steps_per_second": 0.374,
       "step": 7
     },
     {
       "epoch": 8.0,
-      "eval_loss": 0.5233017802238464,
-      "eval_runtime": 2.6981,
-      "eval_samples_per_second": 17.79,
-      "eval_steps_per_second": 0.371,
+      "eval_loss": 0.4572071433067322,
+      "eval_runtime": 2.6696,
+      "eval_samples_per_second": 18.73,
+      "eval_steps_per_second": 0.375,
       "step": 8
     },
     {
       "epoch": 9.0,
-      "eval_loss": 0.4948171079158783,
-      "eval_runtime": 2.6682,
-      "eval_samples_per_second": 17.99,
-      "eval_steps_per_second": 0.375,
+      "eval_loss": 0.4221082031726837,
+      "eval_runtime": 2.677,
+      "eval_samples_per_second": 18.677,
+      "eval_steps_per_second": 0.374,
       "step": 9
     },
     {
       "epoch": 10.0,
-      "grad_norm": 4.900067571807948,
+      "grad_norm": 3.0214434738231213,
       "learning_rate": 0.0,
-      "loss": 0.5927,
+      "loss": 0.5493,
       "step": 10
     },
     {
       "epoch": 10.0,
-      "eval_loss": 0.48538458347320557,
-      "eval_runtime": 2.6954,
-      "eval_samples_per_second": 17.808,
-      "eval_steps_per_second": 0.371,
+      "eval_loss": 0.41154685616493225,
+      "eval_runtime": 2.6558,
+      "eval_samples_per_second": 18.827,
+      "eval_steps_per_second": 0.377,
       "step": 10
     },
     {
       "epoch": 10.0,
       "step": 10,
       "total_flos": 8375186227200.0,
-      "train_loss": 0.8424163699150086,
-      "train_runtime": 304.1051,
-      "train_samples_per_second": 1.578,
-      "train_steps_per_second": 0.033
+      "train_loss": 0.8207268595695496,
+      "train_runtime": 309.7037,
+      "train_samples_per_second": 1.614,
+      "train_steps_per_second": 0.032
     }
   ],
   "logging_steps": 5,