Model save

README.md

@@ -1,7 +1,7 @@
 ---
 base_model: RyanYr/reflect_llama8B_om2-mixed-t0-mstlrg-300k460k-t12_llama33-130k-t12_sft-t1_lr1e-6
 library_name: transformers
-model_name: reflect_llama8B_om2-mstlrg300k460k-llama3370b130k-t12_sft-t1_psdp-t1
+model_name: reflect_llama8B_om2-mstlrg300k460k-llama3370b130k-t12_sft-t1_psdp-t1_b.5
 tags:
 - generated_from_trainer
 - trl
@@ -9,7 +9,7 @@ tags:
 licence: license
 ---
 
-# Model Card for reflect_llama8B_om2-mstlrg300k460k-llama3370b130k-t12_sft-t1_psdp-t1
+# Model Card for reflect_llama8B_om2-mstlrg300k460k-llama3370b130k-t12_sft-t1_psdp-t1_b.5
 
 This model is a fine-tuned version of [RyanYr/reflect_llama8B_om2-mixed-t0-mstlrg-300k460k-t12_llama33-130k-t12_sft-t1_lr1e-6](https://huggingface.co/RyanYr/reflect_llama8B_om2-mixed-t0-mstlrg-300k460k-t12_llama33-130k-t12_sft-t1_lr1e-6).
 It has been trained using [TRL](https://github.com/huggingface/trl).
@@ -20,14 +20,14 @@ It has been trained using [TRL](https://github.com/huggingface/trl).
 from transformers import pipeline
 
 question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
-generator = pipeline("text-generation", model="RyanYr/reflect_llama8B_om2-mstlrg300k460k-llama3370b130k-t12_sft-t1_psdp-t1", device="cuda")
+generator = pipeline("text-generation", model="RyanYr/reflect_llama8B_om2-mstlrg300k460k-llama3370b130k-t12_sft-t1_psdp-t1_b.5", device="cuda")
 output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
 print(output["generated_text"])
 ```
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yyr/huggingface/runs/bw7s6its)
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yyr/huggingface/runs/ijeh3hcd)
 
 This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
 

last_checkpoint/model-00001-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:52661d794e4b3374a4da94f37ee69423834b86b0b9c9877a29b52912b3cb782b
+oid sha256:140216e56bce5f3c1b5265aa4cebb27b3866fa967bc8b1f2daf170aa17d3f3e7
 size 4976706864

last_checkpoint/model-00002-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cad7b364989d89a1e26130ddccaef4fa17013ade23743224ecefe037c607ae8e
+oid sha256:f83dc042c4ac7623fd1b6b67ee6c949a4d727f37e9481b3f5ba0ab43c1e4d6e2
 size 4999802720

last_checkpoint/model-00003-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71abf7029fe39f05b96da3f95af12e9ae8be25399be58eb9c257e74b68a13c08
+oid sha256:19ef2851ad9b0832175cd1997e85a06bd6ff2edc134b49335f753c764e142e11
 size 4915916176

last_checkpoint/model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:678e8a275aaed46efad40829e06f57aa878e07c0c0ca83e3e6600840be6d5bd2
+oid sha256:962ef652afb915df5fe185fa4e73b8925dd035646ff2adcb25e34892546ef4ef
 size 1168147000

last_checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edbfdcfc52122517350c61c1c5d74629e08438dbc3263e6d187cae334d667e08
+oid sha256:72a1c53e16ceb78c4646cce496b22384dfe572737186f7ee1b008632c7fedcba
 size 8056