See axolotl config

axolotl version: 0.9.1.post1

base_model: meta-llama/Llama-3.1-8B-Instruct
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer
gradient_accumulation_steps: 2
micro_batch_size: 8
num_epochs: 4
learning_rate: 0.0001
optimizer: adamw_torch_fused
lr_scheduler: cosine
load_in_8bit: false
load_in_4bit: false
adapter: lora
lora_r: 8
lora_alpha: 16
lora_dropout: 0.05
lora_target_modules:
- q_proj
- k_proj
- v_proj
datasets:
- path: /workspace/FinLoRA/data/train/finlora_sentiment_train.jsonl
  type:
    field_instruction: context
    field_output: target
    format: '[INST] {instruction} [/INST]'
    no_input_format: '[INST] {instruction} [/INST]'
val_set_size: 0.02
output_dir: /workspace/FinLoRA/lora/axolotl-output/sentiment_llama_3_1_8b_fp16_r8
sequence_len: 4096
gradient_checkpointing: true
logging_steps: 500
warmup_steps: 10
evals_per_epoch: 4
saves_per_epoch: 1
weight_decay: 0.0
special_tokens:
  pad_token: <|end_of_text|>
deepspeed: deepspeed_configs/zero1.json
bf16: auto
tf32: false
chat_template: llama3
wandb_name: sentiment_llama_3_1_8b_fp16_r8

workspace/FinLoRA/lora/axolotl-output/sentiment_llama_3_1_8b_fp16_r8

This model is a fine-tuned version of meta-llama/Llama-3.1-8B-Instruct on the /workspace/FinLoRA/data/train/finlora_sentiment_train.jsonl dataset. It achieves the following results on the evaluation set:

• Loss: 0.2154

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• num_devices: 3
• gradient_accumulation_steps: 2
• total_train_batch_size: 48
• total_eval_batch_size: 24
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 10
• num_epochs: 4.0

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0.0007	1	3.3646
No log	0.2502	372	0.2299
0.3256	0.5003	744	0.2189
0.15	0.7505	1116	0.2159
0.15	1.0007	1488	0.2168
0.1247	1.2508	1860	0.2143
0.0994	1.5010	2232	0.2185
0.092	1.7512	2604	0.2113
0.092	2.0013	2976	0.2100
0.0872	2.2515	3348	0.2138
0.0722	2.5017	3720	0.2103
0.0702	2.7518	4092	0.2112
0.0702	3.0020	4464	0.2092
0.0695	3.2522	4836	0.2170
0.0621	3.5024	5208	0.2148
0.0606	3.7525	5580	0.2154

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.8.0.dev20250319+cu128
• Datasets 3.5.1
• Tokenizers 0.21.1