ericrisco/llme2_sft_model_rlaif

README.md

@@ -1,56 +1,58 @@
 ---
 base_model: HuggingFaceH4/zephyr-7b-alpha
-library_name: peft
-license: mit
+library_name: transformers
+model_name: sft_zephyr
 tags:
+- generated_from_trainer
 - trl
 - sft
-- generated_from_trainer
-model-index:
-- name: sft_zephyr
-  results: []
+licence: license
 ---
 
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
-
-# sft_zephyr
-
-This model is a fine-tuned version of [HuggingFaceH4/zephyr-7b-alpha](https://huggingface.co/HuggingFaceH4/zephyr-7b-alpha) on an unknown dataset.
+# Model Card for sft_zephyr
 
-## Model description
+This model is a fine-tuned version of [HuggingFaceH4/zephyr-7b-alpha](https://huggingface.co/HuggingFaceH4/zephyr-7b-alpha).
+It has been trained using [TRL](https://github.com/huggingface/trl).
 
-More information needed
+## Quick start
 
-## Intended uses & limitations
+```python
+from transformers import pipeline
 
-More information needed
+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="ericrisco/sft_zephyr", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
 
-## Training and evaluation data
+## Training procedure
 
-More information needed
+[<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/keepcoding/huggingface/runs/auhb327z) 
 
-## Training procedure
 
-### Training hyperparameters
+This model was trained with SFT.
 
-The following hyperparameters were used during training:
-- learning_rate: 0.0002
-- train_batch_size: 8
-- eval_batch_size: 8
-- seed: 42
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: constant
-- num_epochs: 5
+### Framework versions
 
-### Training results
+- TRL: 0.13.0
+- Transformers: 4.48.0
+- Pytorch: 2.5.1+cu121
+- Datasets: 3.2.0
+- Tokenizers: 0.21.0
 
+## Citations
 
 
-### Framework versions
 
-- PEFT 0.13.0
-- Transformers 4.45.1
-- Pytorch 2.4.1+cu121
-- Datasets 3.0.1
-- Tokenizers 0.20.0
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

adapter_config.json

@@ -3,6 +3,8 @@
   "auto_mapping": null,
   "base_model_name_or_path": "HuggingFaceH4/zephyr-7b-alpha",
   "bias": "none",
+  "eva_config": null,
+  "exclude_modules": null,
   "fan_in_fan_out": false,
   "inference_mode": true,
   "init_lora_weights": true,
@@ -11,6 +13,7 @@
   "layers_to_transform": null,
   "loftq_config": {},
   "lora_alpha": 16,
+  "lora_bias": false,
   "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f7384ff1f60536ac493f61d560cb5effd4dbe9caf6b31c61aea973cc532ef851
+oid sha256:5a649fcb6b3ab986d9c7689dba6fe26f83535a7dafdc0fef7b64e37d22ac691d
 size 109069176

runs/Jan13_20-23-36_ebfdfb1bba2e/events.out.tfevents.1736799820.ebfdfb1bba2e.879.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2acdecb83cb92713377acc177877e94cdbe747654f14791abaa83bdd20a6c80
+size 6211

tokenizer.json

@@ -2,7 +2,7 @@
   "version": "1.0",
   "truncation": {
     "direction": "Left",
-    "max_length": 2048,
+    "max_length": 1024,
     "strategy": "LongestFirst",
     "stride": 0
   },

tokenizer_config.json

@@ -37,6 +37,7 @@
   "chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n'  + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "</s>",
+  "extra_special_tokens": {},
   "legacy": true,
   "model_max_length": 1000000000000000019884624838656,
   "pad_token": "</s>",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:14aea89c032f46779ab624e39111fd6826a62c3a0eb4e09e32d2983e63830f17
-size 5496
+oid sha256:8babcfd6f65eb1ad8faeff60060d21e04e46d247570981b7f3eef740222ecf30
+size 5560