metadata
license: apache-2.0
base_model: mistralai/Mistral-7B-v0.1
tags:
- generated_from_trainer
model-index:
- name: mistral-7B-MedText-epochs-3-lr-0002
results: []
See axolotl config
axolotl version: 0.4.0
base_model: mistralai/Mistral-7B-v0.1
model_type: MistralForCausalLM
tokenizer_type: LlamaTokenizer
is_mistral_derived_model: true
load_in_8bit: false
load_in_4bit: false
strict: false
datasets:
- path: utrgvseniorproject/medtext
type: completion
dataset_prepared_path: last_run_prepared
val_set_size: 0.05
output_dir: ./mistral-7B-MedText-epochs-3-lr-0002
sequence_len: 4096
sample_packing: true
pad_to_sequence_len: true
adapter:
lora_model_dir:
lora_r:
lora_alpha:
lora_dropout:
lora_target_linear:
lora_fan_in_fan_out:
wandb_project: mistral-7B-MedText
wandb_entity: utrgvmedai
wandb_watch:
wandb_name: mistral-7B-MedText-epochs-3-lr-0002
wandb_log_model:
gradient_accumulation_steps: 1
micro_batch_size: 1
num_epochs: 3
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002
train_on_inputs: true
group_by_length: false
bf16: auto
fp16:
tf32: false
gradient_checkpointing: true
early_stopping_patience:
#resume_from_checkpoint: true
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true
flash_attn_cross_entropy: false
flash_attn_rms_norm: true
flash_attn_fuse_qkv: false
flash_attn_fuse_mlp: true
warmup_steps: 100
evals_per_epoch: 4
eval_table_size:
eval_sample_packing: False
saves_per_epoch: 1
debug:
deepspeed: /home/josegomez15/axolotl/deepspeed_configs/zero2.json
weight_decay: 0.1
fsdp:
fsdp_config:
special_tokens:
mistral-7B-MedText-epochs-3-lr-0002
This model is a fine-tuned version of mistralai/Mistral-7B-v0.1 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 7.2922
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- total_train_batch_size: 4
- total_eval_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.3985 | 0.01 | 1 | 1.5677 |
| 1.4776 | 0.25 | 22 | 1.8568 |
| 10.1246 | 0.51 | 44 | 8.7590 |
| 8.1284 | 0.76 | 66 | 8.0049 |
| 7.3967 | 1.01 | 88 | 7.4614 |
| 7.2567 | 1.23 | 110 | 7.2993 |
| 7.3329 | 1.48 | 132 | 7.3749 |
| 7.0671 | 1.74 | 154 | 7.3365 |
| 7.4786 | 1.99 | 176 | 7.3194 |
| 7.3548 | 2.22 | 198 | 7.3092 |
| 7.1782 | 2.47 | 220 | 7.2964 |
| 7.2729 | 2.72 | 242 | 7.2922 |
Framework versions
- Transformers 4.38.0
- Pytorch 2.0.1+cu117
- Datasets 2.17.0
- Tokenizers 0.15.0