Model save

README.md

@@ -3,16 +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
-- alignment-handbook
-- generated_from_trainer
 datasets:
-- scottsuk0306/DepthQA
+- generator
 model-index:
 - name: zephyr-7b-stem-test
   results: []
@@ -23,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-stem-test
 
-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: 0.4085
 
 ## Model description
 
@@ -62,14 +57,15 @@ 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          |
+| 1.1097        | 2.0   | 2    | 1.2839          |
+| 1.1097        | 3.0   | 3    | 1.0813          |
+| 1.1097        | 4.0   | 4    | 0.8745          |
+| 1.0879        | 5.0   | 5    | 0.7450          |
+| 1.0879        | 6.0   | 6    | 0.6025          |
+| 1.0879        | 7.0   | 7    | 0.5167          |
+| 1.0879        | 8.0   | 8    | 0.4539          |
+| 1.0879        | 9.0   | 9    | 0.4191          |
+| 0.547         | 10.0  | 10   | 0.4085          |
 
 
 ### Framework versions

all_results.json

@@ -1,14 +1,9 @@
 {
     "epoch": 10.0,
-    "eval_loss": 0.41154685616493225,
-    "eval_runtime": 3.2821,
-    "eval_samples": 848,
-    "eval_samples_per_second": 15.234,
-    "eval_steps_per_second": 0.305,
     "total_flos": 8375186227200.0,
-    "train_loss": 0.0,
-    "train_runtime": 1.4567,
+    "train_loss": 0.8196125388145447,
+    "train_runtime": 303.2537,
     "train_samples": 848,
-    "train_samples_per_second": 343.244,
-    "train_steps_per_second": 6.865
+    "train_samples_per_second": 1.649,
+    "train_steps_per_second": 0.033
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 10.0,
     "total_flos": 8375186227200.0,
-    "train_loss": 0.0,
-    "train_runtime": 1.4567,
+    "train_loss": 0.8196125388145447,
+    "train_runtime": 303.2537,
     "train_samples": 848,
-    "train_samples_per_second": 343.244,
-    "train_steps_per_second": 6.865
+    "train_samples_per_second": 1.649,
+    "train_steps_per_second": 0.033
 }

trainer_state.json

@@ -10,7 +10,7 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "grad_norm": 17.810282459441467,
+      "grad_norm": 17.81055096288918,
       "learning_rate": 1e-05,
       "loss": 1.1097,
       "step": 1
@@ -18,97 +18,105 @@
     {
       "epoch": 1.0,
       "eval_loss": 1.1095243692398071,
-      "eval_runtime": 2.8023,
-      "eval_samples_per_second": 17.843,
-      "eval_steps_per_second": 0.357,
+      "eval_runtime": 2.808,
+      "eval_samples_per_second": 17.806,
+      "eval_steps_per_second": 0.356,
       "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": 1.2839224338531494,
+      "eval_runtime": 2.6721,
+      "eval_samples_per_second": 18.712,
+      "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": 1.081336259841919,
+      "eval_runtime": 2.6698,
+      "eval_samples_per_second": 18.728,
+      "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_loss": 0.8745010495185852,
+      "eval_runtime": 2.6731,
+      "eval_samples_per_second": 18.705,
       "eval_steps_per_second": 0.374,
       "step": 4
     },
     {
       "epoch": 5.0,
-      "grad_norm": 18.20399999295951,
+      "grad_norm": 18.406465212856943,
       "learning_rate": 5.8682408883346535e-06,
-      "loss": 1.0878,
+      "loss": 1.0879,
       "step": 5
     },
     {
       "epoch": 5.0,
-      "eval_loss": 0.7459021210670471,
-      "eval_runtime": 2.6898,
-      "eval_samples_per_second": 18.589,
-      "eval_steps_per_second": 0.372,
+      "eval_loss": 0.7450248599052429,
+      "eval_runtime": 2.6675,
+      "eval_samples_per_second": 18.744,
+      "eval_steps_per_second": 0.375,
       "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": 0.6025453209877014,
+      "eval_runtime": 2.6697,
+      "eval_samples_per_second": 18.729,
+      "eval_steps_per_second": 0.375,
       "step": 6
     },
     {
       "epoch": 7.0,
-      "eval_loss": 0.5201848745346069,
-      "eval_runtime": 2.6772,
-      "eval_samples_per_second": 18.676,
-      "eval_steps_per_second": 0.374,
+      "eval_loss": 0.5167475938796997,
+      "eval_runtime": 2.6908,
+      "eval_samples_per_second": 18.582,
+      "eval_steps_per_second": 0.372,
       "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": 0.45393621921539307,
+      "eval_runtime": 2.6892,
+      "eval_samples_per_second": 18.593,
+      "eval_steps_per_second": 0.372,
       "step": 8
     },
     {
       "epoch": 9.0,
-      "eval_loss": 0.4221082031726837,
-      "eval_runtime": 2.677,
-      "eval_samples_per_second": 18.677,
-      "eval_steps_per_second": 0.374,
+      "eval_loss": 0.4190720021724701,
+      "eval_runtime": 2.6866,
+      "eval_samples_per_second": 18.611,
+      "eval_steps_per_second": 0.372,
       "step": 9
     },
     {
       "epoch": 10.0,
-      "grad_norm": 3.0214434738231213,
+      "grad_norm": 2.9824311781074426,
       "learning_rate": 0.0,
-      "loss": 0.5493,
+      "loss": 0.547,
+      "step": 10
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 0.408511757850647,
+      "eval_runtime": 2.6822,
+      "eval_samples_per_second": 18.641,
+      "eval_steps_per_second": 0.373,
       "step": 10
     },
     {
       "epoch": 10.0,
       "step": 10,
       "total_flos": 8375186227200.0,
-      "train_loss": 0.0,
-      "train_runtime": 1.4567,
-      "train_samples_per_second": 343.244,
-      "train_steps_per_second": 6.865
+      "train_loss": 0.8196125388145447,
+      "train_runtime": 303.2537,
+      "train_samples_per_second": 1.649,
+      "train_steps_per_second": 0.033
     }
   ],
   "logging_steps": 5,