Model save

README.md

@@ -6,35 +6,30 @@ tags:
 - generated_from_trainer
 - trl
 - sft
-license: apache-2.0
-datasets:
-- OpenAssistant/oasst1
-- allenai/c4
+licence: license
 ---
 
-# notHumpback-M1
+# Model Card for notHumpback-M1
 
-This model follows the Humpback architecture, proposed in the paper [Self-Alignment with Instruction Backtranslation](https://arxiv.org/pdf/2308.06259) 
-by Li et al.
+This model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
 
-It represents the resulting model after the first iteration of self-curation, which is trained on a small amount of gold data 
-and a set of generated data curated by the ["seed model"](https://huggingface.co/Alepach/notHumpback-M0). 
+```python
+from transformers import pipeline
 
-This model can be used for instruction-following.
-It may also be used to, again, score the instruction-response pairs 
-generated by the ["backward model"](https://huggingface.co/Alepach/notHumpback-Myx) for a second iteration of self-curation. 
+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="Alepach/notHumpback-M1", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
 
-Humpback uses instruction backtranslation on a web corpus to generate input-output pairs (self-augmentation), 
-creating a richer dataset for fine-tuning models without the need for additional manual annotation.
-The model then iteratively curates the created dataset, scoring the pairs by quality, and is then finetuned on the resulting subset 
-of all pairs with the highest possible score (self-curation).
+## Training procedure
 
-Varying from the original paper, this model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B).
-It has been trained using [TRL](https://github.com/huggingface/trl).
 
-The dataset used to train this model is a combination of data sampled from the [oasst1](https://huggingface.co/datasets/OpenAssistant/oasst1) 
-dataset and the synthetic dataset which was mentioned above. The latter has been created by applying self-augmentation and self-curation 
-on 502k entries from the english subset ("en") of the [c4](https://huggingface.co/datasets/allenai/c4) dataset.
+
+This model was trained with SFT.
 
 ### Framework versions
 
@@ -46,18 +41,7 @@ on 502k entries from the english subset ("en") of the [c4](https://huggingface.c
 
 ## Citations
 
-Original paper:
 
-```bibtex
-@misc{li2023selfalignment,
-    title={Self-Alignment with Instruction Backtranslation},
-    author={Xian Li and Ping Yu and Chunting Zhou and Timo Schick and Luke Zettlemoyer and Omer Levy and Jason Weston and Mike Lewis},
-    year={2023},
-    eprint={2308.06259},
-    archivePrefix={arXiv},
-    primaryClass={cs.CL}
-}
-```
 
 Cite TRL as:
     

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:781956293cc99f11c53f79fc56a25cc694a398ecaf01e50cd0f50bc4197f4e94
+oid sha256:c4041a510e523ecf82dc5c5d009b232972f4b94e9eae5181a8712140e523c0df
 size 4965799096

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3156b918a65412939a07344f082a227b5451165629df0e570dc143c4b0e0feb7
+oid sha256:dda58e22690fac7ee3929d8165d426ac123f1aae9d7b879e5f49e8f554e7dc83
 size 1459729952

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
-size 17209920
+oid sha256:76cfe2f054560aae896b2b75e273dc97a39e304d4ad19c44a9727a1d6b33c4cc
+size 17210021

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:695df2cc7631d8878a89abb1c77c4fef5bea2dcd14bbc35249581c9dff56d679
+oid sha256:771a5f073e3f38bf66373390370b38d494e2912de8713762f2e145aa1d0cad04
 size 5560