See axolotl config

axolotl version: 0.7.0

adapter: qlora
base_model: meta-llama/Llama-2-7b-hf
bf16: auto
dataset_prepared_path: null
datasets:
- path: mohit9999/all_news_finance_sm_1h2023_custom
  type: alpaca
debug: null
deepspeed: null
early_stopping_patience: null
eval_sample_packing: true
eval_table_size: null
evals_per_epoch: 1
flash_attention: false
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 2
gradient_checkpointing: true
group_by_length: false
hub_model_id: mohit9999/all_news_finance_sm_1h2023_custom_model
learning_rate: 2e-5
load_in_4bit: true
load_in_8bit: false
logging_steps: 1
lora_alpha: 16
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_modules_to_save:
- embed_tokens
- lm_head
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
max_steps: 25
micro_batch_size: 1
model_type: LlamaForCausalLM
num_epochs: 2
optimizer: paged_adamw_8bit
output_dir: ./outputs/lora-out
pad_to_sequence_len: true
resume_from_checkpoint: null
sample_packing: true
saves_per_epoch: 1
sdp_attention: true
sequence_len: 512
special_tokens: null
strict: false
tf32: false
tokenizer_type: LlamaTokenizer
train_on_inputs: false
val_set_size: 0.05
wandb_entity: null
wandb_log_model: null
wandb_name: null
wandb_project: null
wandb_watch: null
warmup_steps: 1
weight_decay: 0.0
xformers_attention: null

all_news_finance_sm_1h2023_custom_model

This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on the mohit9999/all_news_finance_sm_1h2023_custom dataset. It achieves the following results on the evaluation set:

• Loss: 3.7215

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-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• distributed_type: multi-GPU
• num_devices: 8
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• total_eval_batch_size: 8
• optimizer: Use paged_adamw_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 2
• training_steps: 25

Training results

Training Loss	Epoch	Step	Validation Loss
4.3316	0.0150	1	3.8427
4.0068	0.1955	13	3.7215

Framework versions

• PEFT 0.14.0
• Transformers 4.48.3
• Pytorch 2.6.0+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0