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-2-lr-000002

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-2-lr-000002
wandb_log_model:

gradient_accumulation_steps: 1
micro_batch_size: 1
num_epochs: 2
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.000002

train_on_inputs: true
group_by_length: false
bf16: auto
fp16:
tf32: false

gradient_checkpointing: false
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-2-lr-000002

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: 1.5485

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: 2e-06
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 8
• total_eval_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
1.5029	0.02	1	1.5677
1.5889	0.26	11	1.5675
1.2972	0.51	22	1.5647
1.6404	0.77	33	1.5587
1.4796	1.02	44	1.5533
1.428	1.23	55	1.5506
1.5635	1.49	66	1.5492
1.2457	1.74	77	1.5485

Framework versions

• Transformers 4.38.0
• Pytorch 2.0.1+cu117
• Datasets 2.17.0
• Tokenizers 0.15.0