axolotl version: 0.4.0
base_model: BlouseJury/Mistral-7B-Discord-0.1
model_type: MistralForCausalLM
tokenizer_type: LlamaTokenizer
is_mistral_derived_model: true
load_in_8bit: false
load_in_4bit: false
strict: false
datasets:
- path: Intel/orca_dpo_pairs
type:
system_prompt: ""
field_system: system
field_instruction: question
field_output: rejected
field_output: chosen
format: "[INST] {instruction} [/INST]"
no_input_format: "[INST] {instruction} [/INST]"
dataset_prepared_path:
val_set_size: 0.05
output_dir: ./out
sequence_len: 8192
sample_packing: true
pad_to_sequence_len: true
eval_sample_packing: false
wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:
gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 1
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.000005
train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false
gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true
warmup_steps: 10
evals_per_epoch: 4
eval_table_size:
eval_table_max_new_tokens: 128
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
bos_token: "<s>"
eos_token: "</s>"
unk_token: "<unk>"
BlouseJury/Mistral-7B-Discord-0.1-DPO
This model is a fine-tuned version of BlouseJury/Mistral-7B-Discord-0.1 on the Intel/orca_dpo_pairs dataset. It achieves the following results on the evaluation set:
- Loss: 0.7923
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-06
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 10
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.1157 | 0.01 | 1 | 1.1924 |
| 1.0146 | 0.26 | 19 | 0.8381 |
| 0.9004 | 0.51 | 38 | 0.8015 |
| 0.8425 | 0.77 | 57 | 0.7923 |
Framework versions
- Transformers 4.37.0
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 62.29 |
| AI2 Reasoning Challenge (25-Shot) | 63.23 |
| HellaSwag (10-Shot) | 83.27 |
| MMLU (5-Shot) | 62.62 |
| TruthfulQA (0-shot) | 55.28 |
| Winogrande (5-shot) | 78.93 |
| GSM8k (5-shot) | 30.40 |