See axolotl config

axolotl version: 0.4.0

base_model: google/gemma-7b-it
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
tokenizer_config: philschmid/gemma-tokenizer-chatml
tokenizer_use_fast: true

load_in_8bit: false
load_in_4bit: true
strict: false

chat_template: chatml
datasets:
  - path: manishiitg/aditi-syn-train-small
    type: completion
  
hub_model_id: manishiitg/open-aditi-chat-hi-1.96-gemma
hf_use_auth_token: true

wandb_project: open-aditi-chat-hi-1.96-gemma

dataset_prepared_path: manishiitg
push_dataset_to_hub: manishiitg
val_set_size: .1
output_dir: /sky-notebook/manishiitg/open-aditi-chat-hi-1.96-gemma

adapter: qlora
lora_model_dir:
save_safetensors: true

sequence_len: 4096
sample_packing: true
pad_to_sequence_len: true
eval_sample_packing: false

lora_r: 64
lora_alpha: 32
lora_dropout: 0.05
lora_target_linear: true

wandb_entity:
wandb_watch:
wandb_run_id:
wandb_log_model:

gradient_accumulation_steps: 8
micro_batch_size: 1
num_epochs: 2
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002

adam_beta2: 0.95
adam_epsilon: 0.00001
max_grad_norm: 1.0

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


gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
auto_resume_from_checkpoints: true ## manage check point resume from here
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
save_steps: 20 ## increase based on your dataset
save_strategy: steps
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:

open-aditi-chat-hi-1.96-gemma

This model is a fine-tuned version of google/gemma-7b-it on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.3650

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: 8
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• total_eval_batch_size: 8
• optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-05
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss
4.1948	0.01	1	6.3159
1.6405	0.25	25	1.8789
1.3525	0.5	50	1.6130
1.2195	0.75	75	1.4971
1.2072	1.0	100	1.4285
1.1556	1.23	125	1.3897
1.1202	1.48	150	1.3726
1.144	1.73	175	1.3650

Framework versions

• PEFT 0.9.0
• Transformers 4.40.0.dev0
• Pytorch 2.1.2+cu121
• Datasets 2.18.0
• Tokenizers 0.15.0