Training in progress, step 100

README.md

@@ -1,71 +1,102 @@
 ---
-library_name: peft
+license: mit
+library_name: "trl"
 tags:
-- trl
-- kto
-- generated_from_trainer
+- KTO
+- WeniGPT
 base_model: HuggingFaceH4/zephyr-7b-beta
 model-index:
-- name: WeniGPT-Agents-Zephyr-1.0.27-KTO
+- name: Weni/WeniGPT-Agents-Zephyr-1.0.27-KTO
   results: []
+language: ['pt']
 ---
 
-<!-- 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. -->
+# Weni/WeniGPT-Agents-Zephyr-1.0.27-KTO
 
-# WeniGPT-Agents-Zephyr-1.0.27-KTO
+This model is a fine-tuned version of [HuggingFaceH4/zephyr-7b-beta] on the dataset Weni/wenigpt-agent-1.4.0 with the KTO trainer. It is part of the WeniGPT project for [Weni](https://weni.ai/).
+Description: Experiment with a new tokenizer configuration for chat template of zephyr
 
-This model is a fine-tuned version of [HuggingFaceH4/zephyr-7b-beta](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.4799
-- Rewards/chosen: 0.0648
-- Rewards/rejected: -0.1020
-- Rewards/margins: 0.1668
-- Kl: 0.9039
-- Logps/chosen: -278.8387
-- Logps/rejected: -239.1956
+{'eval_loss': 0.47991228103637695, 'eval_runtime': 169.8862, 'eval_samples_per_second': 2.06, 'eval_steps_per_second': 0.518, 'eval_rewards/chosen': 0.06480830907821655, 'eval_rewards/rejected': -0.10202363133430481, 'eval_rewards/margins': 0.16683195531368256, 'eval_kl': 0.9039192199707031, 'eval_logps/chosen': -278.8386535644531, 'eval_logps/rejected': -239.19561767578125, 'epoch': 0.97}
 
-## Model description
+## Intended uses & limitations
 
-More information needed
+This model has not been trained to avoid specific intructions. 
 
-## Intended uses & limitations
+## Training procedure
 
-More information needed
+Finetuning was done on the model HuggingFaceH4/zephyr-7b-beta with the following prompt:
 
-## Training and evaluation data
+```
+---------------------
+System_prompt:
+Agora você se chama {name}, você é {occupation} e seu objetivo é {chatbot_goal}. O adjetivo que mais define a sua personalidade é {adjective} e você se comporta da seguinte forma:
+{instructions_formatted}
 
-More information needed
+Na sua memória você tem esse contexto:
+{context}
+
+Lista de requisitos:
+ - Responda de forma natural, mas nunca fale sobre um assunto fora do contexto.
+ - Nunca traga informações do seu próprio conhecimento.
+ - Repito é crucial que você responda usando apenas informações do contexto.
+ - Nunca mencione o contexto fornecido.
+ - Nunca mencione a pergunta fornecida.
+ - Gere a resposta mais útil possível para a pergunta usando informações do conexto acima.
+ - Nunca elabore sobre o porque e como você fez a tarefa, apenas responda.
+
+
+---------------------
+Question:
+{question}
 
-## Training procedure
+
+---------------------
+Response:
+{answer}
+
+
+---------------------
+
+```
 
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-06
-- train_batch_size: 4
-- eval_batch_size: 4
-- seed: 42
+- per_device_train_batch_size: 4
+- per_device_eval_batch_size: 4
 - gradient_accumulation_steps: 4
+- num_gpus: 1
 - total_train_batch_size: 16
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: linear
-- lr_scheduler_warmup_ratio: 0.03
-- training_steps: 147
-- mixed_precision_training: Native AMP
+- optimizer: AdamW
+- lr_scheduler_type: cosine
+- num_steps: 147
+- quantization_type: bitsandbytes
+- LoRA: ("\n  - bits: 4\n  - use_exllama: True\n  - device_map: auto\n  - use_cache: False\n  - lora_r: 8\n  - lora_alpha: 16\n  - lora_dropout: 0.05\n  - bias: none\n  - target_modules: ['q_proj', 'k_proj', 'v_proj', 'o_proj', 'gate_proj', 'up_proj', 'down_proj', 'lm_head', 'embed_tokens']\n  - task_type: CAUSAL_LM",)
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/margins | Kl     | Logps/chosen | Logps/rejected |
-|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:---------------:|:------:|:------------:|:--------------:|
-| 0.5716        | 0.33  | 50   | 0.4898          | 0.0297         | -0.0529          | 0.0826          | 0.4425 | -279.1900    | -238.7043      |
-| 0.6643        | 0.66  | 100  | 0.4799          | 0.0648         | -0.1020          | 0.1668          | 0.9039 | -278.8387    | -239.1956      |
-
-
 ### Framework versions
 
-- PEFT 0.10.0
-- Transformers 4.38.2
-- Pytorch 2.1.0+cu118
-- Datasets 2.18.0
-- Tokenizers 0.15.2
+- transformers==4.38.2
+- datasets==2.18.0
+- peft==0.10.0
+- safetensors==0.4.2
+- evaluate==0.4.1
+- bitsandbytes==0.43
+- huggingface_hub==0.22.2
+- seqeval==1.2.2
+- optimum==1.18.1
+- auto-gptq==0.7.1
+- gpustat==1.1.1
+- deepspeed==0.14.0
+- wandb==0.16.6
+- trl==0.8.1
+- accelerate==0.29.2
+- coloredlogs==15.0.1
+- traitlets==5.14.2
+- autoawq@https://github.com/casper-hansen/AutoAWQ/releases/download/v0.2.4/autoawq-0.2.4+cu118-cp310-cp310-linux_x86_64.whl
+
+### Hardware
+- Cloud provided: runpod.io

adapter_config.json

@@ -20,15 +20,15 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "up_proj",
-    "embed_tokens",
-    "down_proj",
-    "k_proj",
-    "gate_proj",
     "q_proj",
     "v_proj",
+    "down_proj",
+    "up_proj",
+    "k_proj",
+    "lm_head",
     "o_proj",
-    "lm_head"
+    "embed_tokens",
+    "gate_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:510ebbd68ad2fc6e65f885d86a5b827f90c99eeb99b27ae8facd47869b266a3b
+oid sha256:cd57ab79ef124274d11e15da7954eead2132774d0e7119df267cce5cbd1d1105
 size 1134834064

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a423771de9ce0a2d017dee74ab431281f61b41fda09615f641d9c7ad58010a7
+oid sha256:1bddc716402039400132fe7ed2438b8ffa438d294fae7f94ce5c8c97e9abe6e6
 size 5688