dosa777/llama3-1b

See axolotl config

axolotl version: 0.8.0.dev0

base_model: NousResearch/Hermes-3-Llama-3.2-3B
# optionally might have model_type or tokenizer_type
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer
# Automatically upload checkpoint and final model to HF
# hub_model_id: username/custom_model_name

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:
lora_modules_to_save:
  - embed_tokens
  - lm_head

wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

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
s2_attention:

warmup_steps: 10
evals_per_epoch: 4
eval_table_size:
eval_max_new_tokens: 128
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
   pad_token: <|end_of_text|>

outputs/lora-out

This model is a fine-tuned version of NousResearch/Hermes-3-Llama-3.2-3B on the mhenrichsen/alpaca_2k_test dataset. It achieves the following results on the evaluation set:

• Loss: 0.9717

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: Use OptimizerNames.ADAMW_BNB 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
1.2051	0.0930	1	1.3039
1.0777	0.2791	3	1.2884
1.1398	0.5581	6	1.1808
1.0909	0.8372	9	1.0715
0.8999	1.0930	12	0.9657
0.7534	1.3721	15	0.9733
0.7596	1.6512	18	0.9730
0.7925	1.9302	21	0.9712
0.6335	2.1860	24	0.9683
0.6694	2.4651	27	0.9722
0.6156	2.7442	30	0.9713
0.6567	3.0	33	0.9704
0.6581	3.2791	36	0.9713
0.6023	3.5581	39	0.9717

Framework versions

• PEFT 0.14.0
• Transformers 4.49.0
• Pytorch 2.5.1+cu124
• Datasets 3.2.0
• Tokenizers 0.21.0