base_model: meta-llama/Llama-3.2-1B-Instruct
hub_model_id: minpeter/llama-1b-tool-vllm-ci

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  # 0.5k
  - path: minpeter/xlam-function-calling-60k-hermes
    data_files:
      - result.parquet
    type: chat_template
    chat_template: llama3
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    shards: 120

  # 0.35k
  - path: minpeter/xlam-irrelevance-7.5k-qwen2.5-72b-distill-hermes
    data_files:
      - result.parquet
    type: chat_template
    chat_template: llama3
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    shards: 15

  #  1.2k
  - path: minpeter/hermes-function-calling-v1-jsonl
    data_files:
      - func-calling-singleturn.jsonl
      - func-calling.jsonl
    type: chat_template
    chat_template: llama3
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    shards: 3

  # 1k
  - path: minpeter/apigen-mt-5k-friendli
    data_files:
      - train.jsonl
    type: chat_template
    chat_template: llama3
    field_messages: messages
    message_field_role: role
    message_field_content: content
    shards: 5

chat_template: llama3

dataset_prepared_path: last_run_prepared

output_dir: ./output

adapter: lora
lora_model_dir:

sequence_len: 4096
pad_to_sequence_len: true
sample_packing: true

val_set_size: 0.05
eval_sample_packing: true
evals_per_epoch: 3

lora_r: 8
lora_alpha: 16
lora_dropout: 0.05
lora_fan_in_fan_out:
lora_target_modules:
  - gate_proj
  - down_proj
  - up_proj
  - q_proj
  - v_proj
  - k_proj
  - o_proj

wandb_project: "axolotl"
wandb_entity: "kasfiekfs-e"
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 2
micro_batch_size: 2
num_epochs: 2
optimizer: adamw_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

loss_watchdog_threshold: 5.0
loss_watchdog_patience: 3

warmup_steps: 10
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:

special_tokens:
  pad_token: "<|finetune_right_pad_id|>"

llama-1b-tool-vllm-ci

This model is a fine-tuned version of meta-llama/Llama-3.2-1B-Instruct on the minpeter/xlam-function-calling-60k-hermes, the minpeter/xlam-irrelevance-7.5k-qwen2.5-72b-distill-hermes, the minpeter/hermes-function-calling-v1-jsonl and the minpeter/apigen-mt-5k-friendli datasets. It achieves the following results on the evaluation set:

Loss: 0.4122

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: 2
total_train_batch_size: 4
optimizer: Use OptimizerNames.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: 10
training_steps: 380

Training results

Training Loss	Epoch	Step	Validation Loss
No log	0	0	0.9339
0.3676	0.3360	64	0.4842
0.493	0.6719	128	0.4455
0.4702	1.0052	192	0.4302
0.574	1.3412	256	0.4195
0.4548	1.6772	320	0.4122

Framework versions

PEFT 0.15.2
Transformers 4.52.4
Pytorch 2.6.0+cu124
Datasets 3.6.0
Tokenizers 0.21.1

minpeter/LoRA-Llama-3.2-1B-tool-vllm-ci

llama-1b-tool-vllm-ci

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

minpeter/
LoRA-Llama-3.2-1B-tool-vllm-ci