Model save

README.md

@@ -3,15 +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:
-- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: zephyr-7b-align-scan-0.0-0.0-cosine-2
   results: []
@@ -22,17 +17,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-align-scan-0.0-0.0-cosine-2
 
-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 HuggingFaceH4/ultrafeedback_binarized 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 an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6691
-- Rewards/chosen: -0.3145
-- Rewards/rejected: -0.5033
-- Rewards/accuracies: 0.3075
-- Rewards/margins: 0.1888
-- Logps/rejected: -601.5961
-- Logps/chosen: -399.7386
-- Logits/rejected: 4.1594
-- Logits/chosen: 3.9982
+- Logits/chosen: 2.2227
+- Logits/rejected: 2.3998
+- Logps/chosen: -231.4086
+- Logps/rejected: -327.2254
+- Loss: 0.6710
+- Rewards/accuracies: 0.3155
+- Rewards/chosen: -0.1518
+- Rewards/margins: 0.0862
+- Rewards/rejected: -0.2380
 
 ## Model description
 
@@ -51,7 +46,7 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 8.932373406244385e-07
+- learning_rate: 3.454537113325051e-07
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
@@ -67,9 +62,9 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
-|:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
-| 0.66          | 1.0417 | 100  | 0.6710          | -0.1518        | -0.2380          | 0.3155             | 0.0862          | -327.2254      | -231.4086    | 2.3998          | 2.2227        |
+| Training Loss | Epoch  | Step | Logits/chosen | Logits/rejected | Logps/chosen | Logps/rejected | Validation Loss | Rewards/accuracies | Rewards/chosen | Rewards/margins | Rewards/rejected |
+|:-------------:|:------:|:----:|:-------------:|:---------------:|:------------:|:--------------:|:---------------:|:------------------:|:--------------:|:---------------:|:----------------:|
+| 0.66          | 1.0417 | 100  | 2.2227        | 2.3998          | -231.4086    | -327.2254      | 0.6710          | 0.3155             | -0.1518        | 0.0862          | -0.2380          |
 
 
 ### Framework versions

all_results.json

@@ -14,9 +14,9 @@
     "eval_samples_per_second": 17.613,
     "eval_steps_per_second": 0.555,
     "total_flos": 0.0,
-    "train_loss": 0.6542075766871372,
-    "train_runtime": 2362.1376,
+    "train_loss": 0.0,
+    "train_runtime": 0.0151,
     "train_samples": 6113,
-    "train_samples_per_second": 5.176,
-    "train_steps_per_second": 0.081
+    "train_samples_per_second": 807844.742,
+    "train_steps_per_second": 12686.585
 }

config.json

@@ -22,6 +22,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": true,
+  "use_cache": false,
   "vocab_size": 32000
 }

runs/Sep28_05-41-43_qa-a100-005.crc.nd.edu/events.out.tfevents.1727516750.qa-a100-005.crc.nd.edu.103038.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20e9a72ad1adfba4d325330f4f84d263ce178b64b3dd7c8d3ff59bce3cd57385
+size 6808

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 2.0,
     "total_flos": 0.0,
-    "train_loss": 0.6542075766871372,
-    "train_runtime": 2362.1376,
+    "train_loss": 0.0,
+    "train_runtime": 0.0151,
     "train_samples": 6113,
-    "train_samples_per_second": 5.176,
-    "train_steps_per_second": 0.081
+    "train_samples_per_second": 807844.742,
+    "train_steps_per_second": 12686.585
 }

trainer_state.json

@@ -328,10 +328,10 @@
       "epoch": 2.0,
       "step": 192,
       "total_flos": 0.0,
-      "train_loss": 0.6542075766871372,
-      "train_runtime": 2362.1376,
-      "train_samples_per_second": 5.176,
-      "train_steps_per_second": 0.081
+      "train_loss": 0.0,
+      "train_runtime": 0.0151,
+      "train_samples_per_second": 807844.742,
+      "train_steps_per_second": 12686.585
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:62f8749eb732d929518f27ac38e596a1f235270396c09fc8453f000ed47f4d13
+oid sha256:7e72b12738e7223504a07d5d00dad20ac1780fd7771a882714ca5d69e9ccd7ee
 size 7672