Model save

README.md

@@ -2,15 +2,9 @@
 license: mit
 base_model: HuggingFaceH4/mistral-7b-sft-beta
 tags:
-- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
-- trl
-- dpo
-- generated_from_trainer
-datasets:
-- RedMist137/AIHF_DPO_iter0
 model-index:
 - name: DPO-Zephyr-7B
   results: []
@@ -21,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # DPO-Zephyr-7B
 
-This model is a fine-tuned version of [HuggingFaceH4/mistral-7b-sft-beta](https://huggingface.co/HuggingFaceH4/mistral-7b-sft-beta) on the RedMist137/AIHF_DPO_iter0 dataset.
+This model is a fine-tuned version of [HuggingFaceH4/mistral-7b-sft-beta](https://huggingface.co/HuggingFaceH4/mistral-7b-sft-beta) on the None dataset.
 
 ## Model description
 
@@ -46,8 +40,8 @@ The following hyperparameters were used during training:
 - seed: 42
 - distributed_type: multi-GPU
 - num_devices: 8
-- gradient_accumulation_steps: 32
-- total_train_batch_size: 256
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 128
 - total_eval_batch_size: 8
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine

all_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 0.9984,
+    "epoch": 0.9993235625704623,
     "total_flos": 0.0,
-    "train_loss": 0.6750027988201532,
-    "train_runtime": 5038.0071,
-    "train_samples": 40000,
-    "train_samples_per_second": 7.94,
-    "train_steps_per_second": 0.031
+    "train_loss": 0.6513021779835009,
+    "train_runtime": 4537.7195,
+    "train_samples": 35474,
+    "train_samples_per_second": 7.818,
+    "train_steps_per_second": 0.061
 }

model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9b4e1f0f9bc753697fb6587a38ef56b0025bff1fbdf2d1953024d8484eee6827
+oid sha256:f5f179c427e1375727182574bc3a342b4caad2d6cd5f2069b3aa86ec52532fd4
 size 4943162336

model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:764c948955c5f30f06a23eb2b34c04f2d36cf8207e4b16b8d0e11ff808c79b38
+oid sha256:9b624de65f7536e9d6ea0c2e0d27e3c8780ff109ea9e023473b5b772cefe86af
 size 4999819336

model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:78ff5c72795d18b2aa7862fa4b01506849b01b1d859a697afd52da7565212186
+oid sha256:0e548873358e30afd5304f376998d717d3306d3673b19b984a392ff738ead249
 size 4540516344

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 0.9984,
+    "epoch": 0.9993235625704623,
     "total_flos": 0.0,
-    "train_loss": 0.6750027988201532,
-    "train_runtime": 5038.0071,
-    "train_samples": 40000,
-    "train_samples_per_second": 7.94,
-    "train_steps_per_second": 0.031
+    "train_loss": 0.6513021779835009,
+    "train_runtime": 4537.7195,
+    "train_samples": 35474,
+    "train_samples_per_second": 7.818,
+    "train_steps_per_second": 0.061
 }