Model save

README.md

@@ -3,16 +3,10 @@ library_name: transformers
 license: apache-2.0
 base_model: alignment-handbook/zephyr-7b-sft-full
 tags:
-- alignment-handbook
-- trl
-- dpo
-- generated_from_trainer
 - trl
 - dpo
 - alignment-handbook
 - generated_from_trainer
-datasets:
-- data/zephyr_uf_rlced_conifer_ref
 model-index:
 - name: uf-rlced-conifer-zephyr-7b-group-dpo-no-clip-no-excess
   results: []
@@ -23,17 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # uf-rlced-conifer-zephyr-7b-group-dpo-no-clip-no-excess
 
-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 data/zephyr_uf_rlced_conifer_ref dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.3975
-- Rewards/chosen: -1.2128
-- Rewards/rejected: -2.4224
-- Rewards/accuracies: 0.8130
-- Rewards/margins: 1.2096
-- Logps/rejected: -652.3658
-- Logps/chosen: -505.8686
-- Logits/rejected: 0.3206
-- Logits/chosen: -0.1209
+This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on an unknown dataset.
 
 ## Model description
 

all_results.json

@@ -14,9 +14,9 @@
     "eval_samples_per_second": 21.881,
     "eval_steps_per_second": 0.343,
     "total_flos": 0.0,
-    "train_loss": 0.5182241236997975,
-    "train_runtime": 20862.0677,
+    "train_loss": 0.5189602105981774,
+    "train_runtime": 20904.8753,
     "train_samples": 184443,
-    "train_samples_per_second": 8.841,
-    "train_steps_per_second": 0.035
+    "train_samples_per_second": 8.823,
+    "train_steps_per_second": 0.034
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 0.9993060374739764,
     "total_flos": 0.0,
-    "train_loss": 0.5182241236997975,
-    "train_runtime": 20862.0677,
+    "train_loss": 0.5189602105981774,
+    "train_runtime": 20904.8753,
     "train_samples": 184443,
-    "train_samples_per_second": 8.841,
-    "train_steps_per_second": 0.035
+    "train_samples_per_second": 8.823,
+    "train_steps_per_second": 0.034
 }