See axolotl config

axolotl version: 0.4.1

# Upload the final model to Huggingface
hub_model_id: shalini03/tinyllama-1.1B_alpaca_2k_lora

# Store the training logs in weights and biases
#wandb_entity: shalini_03
#wandb_project: ft_tinyllama-1.1B_alpaca_2k_lora

# The rest of this config stays the same:
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
model_type: LlamaForCausalLM
tokenizer_type: LlamaTokenizer

load_in_8bit: true
load_in_4bit: false
strict: false

datasets:
  - path: mhenrichsen/alpaca_2k_test
    type: alpaca
dataset_prepared_path:
val_set_size: 0.05
output_dir: ./outputs/lora-out

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

adapter: lora
lora_model_dir:
lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:

gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 4
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0002

train_on_inputs: false
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

warmup_steps: 10
evals_per_epoch: 4
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:

tinyllama-1.1B_alpaca_2k_lora

This model is a fine-tuned version of TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.2111

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss
1.4615	0.0816	1	1.4899
1.3851	0.2449	3	1.4869
1.3658	0.4898	6	1.4376
1.2683	0.7347	9	1.3399
1.2259	0.9796	12	1.2956
1.2523	1.1633	15	1.2787
1.2271	1.4082	18	1.2527
1.1348	1.6531	21	1.2336
1.2694	1.8980	24	1.2286
1.1484	2.0816	27	1.2224
1.1527	2.3265	30	1.2214
1.1937	2.5714	33	1.2187
1.1121	2.8163	36	1.2150
1.1517	3.0612	39	1.2147
1.1888	3.2449	42	1.2107
1.1002	3.4898	45	1.2122
1.1884	3.7347	48	1.2111

Framework versions

• PEFT 0.11.1
• Transformers 4.41.1
• Pytorch 2.1.2+cu118
• Datasets 2.19.1
• Tokenizers 0.19.1