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: 0.4115
+- Loss: 0.1468
 
 ## Model description
 
@@ -60,16 +56,16 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 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          |
+| 1.0647        | 1.0   | 1    | 1.0615          |
+| 1.0647        | 2.0   | 2    | 1.1875          |
+| 1.0647        | 3.0   | 3    | 0.9856          |
+| 1.0647        | 4.0   | 4    | 0.6196          |
+| 0.969         | 5.0   | 5    | 0.4429          |
+| 0.969         | 6.0   | 6    | 0.3590          |
+| 0.969         | 7.0   | 7    | 0.2403          |
+| 0.969         | 8.0   | 8    | 0.1860          |
+| 0.969         | 9.0   | 9    | 0.1558          |
+| 0.2782        | 10.0  | 10   | 0.1468          |
 
 
 ### Framework versions

all_results.json

@@ -1,14 +1,9 @@
 {
     "epoch": 10.0,
-    "eval_loss": 0.41154685616493225,
-    "eval_runtime": 2.677,
-    "eval_samples": 848,
-    "eval_samples_per_second": 18.678,
-    "eval_steps_per_second": 0.374,
     "total_flos": 8375186227200.0,
-    "train_loss": 0.8207268595695496,
-    "train_runtime": 309.7037,
-    "train_samples": 848,
-    "train_samples_per_second": 1.614,
+    "train_loss": 0.6331865787506104,
+    "train_runtime": 307.8989,
+    "train_samples": 424,
+    "train_samples_per_second": 0.617,
     "train_steps_per_second": 0.032
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 10.0,
     "total_flos": 8375186227200.0,
-    "train_loss": 0.8207268595695496,
-    "train_runtime": 309.7037,
-    "train_samples": 848,
-    "train_samples_per_second": 1.614,
+    "train_loss": 0.6331865787506104,
+    "train_runtime": 307.8989,
+    "train_samples": 424,
+    "train_samples_per_second": 0.617,
     "train_steps_per_second": 0.032
 }

trainer_state.json

@@ -10,112 +10,112 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 17.810282459441467,
+      "grad_norm": 25.3735801083628,
       "learning_rate": 1e-05,
-      "loss": 1.1097,
+      "loss": 1.0647,
       "step": 1
     },
     {
       "epoch": 1.0,
-      "eval_loss": 1.1095243692398071,
-      "eval_runtime": 2.8023,
-      "eval_samples_per_second": 17.843,
+      "eval_loss": 1.061476469039917,
+      "eval_runtime": 2.8031,
+      "eval_samples_per_second": 6.778,
       "eval_steps_per_second": 0.357,
       "step": 1
     },
     {
       "epoch": 2.0,
-      "eval_loss": 1.2839607000350952,
-      "eval_runtime": 2.6692,
-      "eval_samples_per_second": 18.732,
-      "eval_steps_per_second": 0.375,
+      "eval_loss": 1.1875081062316895,
+      "eval_runtime": 2.6772,
+      "eval_samples_per_second": 7.097,
+      "eval_steps_per_second": 0.374,
       "step": 2
     },
     {
       "epoch": 3.0,
-      "eval_loss": 1.0814800262451172,
-      "eval_runtime": 2.6746,
-      "eval_samples_per_second": 18.695,
-      "eval_steps_per_second": 0.374,
+      "eval_loss": 0.985565185546875,
+      "eval_runtime": 2.6682,
+      "eval_samples_per_second": 7.121,
+      "eval_steps_per_second": 0.375,
       "step": 3
     },
     {
       "epoch": 4.0,
-      "eval_loss": 0.874021053314209,
-      "eval_runtime": 2.6724,
-      "eval_samples_per_second": 18.71,
-      "eval_steps_per_second": 0.374,
+      "eval_loss": 0.6195911169052124,
+      "eval_runtime": 2.6829,
+      "eval_samples_per_second": 7.082,
+      "eval_steps_per_second": 0.373,
       "step": 4
     },
     {
       "epoch": 5.0,
-      "grad_norm": 18.20399999295951,
+      "grad_norm": 13.767739837044488,
       "learning_rate": 5.8682408883346535e-06,
-      "loss": 1.0878,
+      "loss": 0.969,
       "step": 5
     },
     {
       "epoch": 5.0,
-      "eval_loss": 0.7459021210670471,
-      "eval_runtime": 2.6898,
-      "eval_samples_per_second": 18.589,
+      "eval_loss": 0.4429333806037903,
+      "eval_runtime": 2.6868,
+      "eval_samples_per_second": 7.072,
       "eval_steps_per_second": 0.372,
       "step": 5
     },
     {
       "epoch": 6.0,
-      "eval_loss": 0.6033510565757751,
-      "eval_runtime": 2.6879,
-      "eval_samples_per_second": 18.602,
-      "eval_steps_per_second": 0.372,
+      "eval_loss": 0.3589998185634613,
+      "eval_runtime": 2.6805,
+      "eval_samples_per_second": 7.088,
+      "eval_steps_per_second": 0.373,
       "step": 6
     },
     {
       "epoch": 7.0,
-      "eval_loss": 0.5201848745346069,
-      "eval_runtime": 2.6772,
-      "eval_samples_per_second": 18.676,
+      "eval_loss": 0.24026596546173096,
+      "eval_runtime": 2.671,
+      "eval_samples_per_second": 7.113,
       "eval_steps_per_second": 0.374,
       "step": 7
     },
     {
       "epoch": 8.0,
-      "eval_loss": 0.4572071433067322,
-      "eval_runtime": 2.6696,
-      "eval_samples_per_second": 18.73,
-      "eval_steps_per_second": 0.375,
+      "eval_loss": 0.18603171408176422,
+      "eval_runtime": 2.7105,
+      "eval_samples_per_second": 7.01,
+      "eval_steps_per_second": 0.369,
       "step": 8
     },
     {
       "epoch": 9.0,
-      "eval_loss": 0.4221082031726837,
-      "eval_runtime": 2.677,
-      "eval_samples_per_second": 18.677,
+      "eval_loss": 0.15581761300563812,
+      "eval_runtime": 2.6748,
+      "eval_samples_per_second": 7.103,
       "eval_steps_per_second": 0.374,
       "step": 9
     },
     {
       "epoch": 10.0,
-      "grad_norm": 3.0214434738231213,
+      "grad_norm": 2.4836614664887184,
       "learning_rate": 0.0,
-      "loss": 0.5493,
+      "loss": 0.2782,
       "step": 10
     },
     {
       "epoch": 10.0,
-      "eval_loss": 0.41154685616493225,
-      "eval_runtime": 2.6558,
-      "eval_samples_per_second": 18.827,
-      "eval_steps_per_second": 0.377,
+      "eval_loss": 0.14684411883354187,
+      "eval_runtime": 2.6767,
+      "eval_samples_per_second": 7.098,
+      "eval_steps_per_second": 0.374,
       "step": 10
     },
     {
       "epoch": 10.0,
       "step": 10,
       "total_flos": 8375186227200.0,
-      "train_loss": 0.8207268595695496,
-      "train_runtime": 309.7037,
-      "train_samples_per_second": 1.614,
+      "train_loss": 0.6331865787506104,
+      "train_runtime": 307.8989,
+      "train_samples_per_second": 0.617,
       "train_steps_per_second": 0.032
     }
   ],