joseagmz/mistral-7B-Tinybook-epochs-6-lr-0002

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/Tinybook
    type: completion
dataset_prepared_path: /home/josegomez15/med-llm/last_run_prepared
val_set_size: 0.05
output_dir: ./mistral-7B-Tinybook-epochs-5-lr-0002

sequence_len: 4096
sample_packing: false
pad_to_sequence_len: true

wandb_project: mistral-7B-Tinybook
wandb_entity: utrgvmedai
wandb_watch:
wandb_name: mistral-7B-Tinybook-epochs-6-lr-0002
wandb_log_model:

gradient_accumulation_steps: 1
micro_batch_size: 16
num_epochs: 6
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002

train_on_inputs: True # make sure you have this on True
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
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:
saves_per_epoch: 1
debug:
deepspeed: /home/josegomez15/axolotl/deepspeed_configs/zero2.json
weight_decay: 0.1
fsdp:
fsdp_config:
special_tokens:

mistral-7B-Tinybook-epochs-5-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: 1.2972

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: 16
• eval_batch_size: 16
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• total_train_batch_size: 128
• total_eval_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss
1.4261	1.0	1	1.5358
1.4254	2.0	2	1.5169
1.3801	3.0	3	1.4326
1.1051	4.0	4	1.5059
0.9669	5.0	5	1.3067
0.7256	6.0	6	1.2972

Framework versions

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