---

Model Test Score Comparison (BFCL)

Test Item	tool Model Accuracy	base Model Accuracy	Score Difference (tool - base)
irrelevance	0.8708	0.4333	+0.4375
multi_turn_base	0.0350	0.0100	+0.0250
parallel_multiple	0.7950	0.4750	+0.3200
parallel	0.7900	0.5200	+0.2700
simple	0.8375	0.7575	+0.0800
multiple	0.8700	0.7650	+0.1050

See axolotl config

axolotl version: 0.10.0.dev0

base_model: minpeter/HyperCLOVAX-SEED-Text-Instruct-3B-hf
hub_model_id: minpeter/LoRA-HCX-3b-sf-xlam-01

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: minpeter/xlam-function-calling-60k-hermes
    data_files:
      - result.parquet
    type: chat_template
    roles_to_train: ["assistant"]
    field_messages: conversations
    message_property_mappings:
      role: from
      content: value
    shards: 120
  - path: minpeter/xlam-irrelevance-7.5k-qwen2.5-72b-distill-hermes
    data_files:
      - result.parquet
    type: chat_template
    roles_to_train: ["assistant"]
    field_messages: conversations
    message_property_mappings:
      role: from
      content: value
    shards: 15
  - path: minpeter/hermes-function-calling-v1-jsonl
    data_files:
      - func-calling-singleturn.jsonl
      - func-calling.jsonl
    type: chat_template
    roles_to_train: ["assistant"]
    field_messages: conversations
    message_property_mappings:
      role: from
      content: value
    shards: 3
  - path: minpeter/hermes-function-calling-v1-jsonl
    data_files:
      - glaive-function-calling-5k.jsonl
    type: chat_template
    roles_to_train: ["assistant"]
    field_messages: conversations
    message_property_mappings:
      role: from
      content: value
    shards: 5
  - path: minpeter/apigen-mt-5k-friendli
    data_files:
      - train.jsonl
      - test.jsonl
    type: chat_template
    roles_to_train: ["assistant"]
    field_messages: messages
    message_property_mappings:
      role: role
      content: content
    shards: 10
chat_template: chatml

dataset_prepared_path: last_run_prepared

output_dir: ./output

adapter: lora
lora_model_dir:

sequence_len: 20000
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:
  bos_token: "<|im_start|>"
  eos_token: "<|im_end|>"
  pad_token: "<|endoftext|>"

LoRA-HCX-3b-sf-xlam-01

This model is a fine-tuned version of minpeter/HyperCLOVAX-SEED-Text-Instruct-3B-hf 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, 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: 0.3609

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
• num_epochs: 2.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.7577	0.0108	1	0.8759
0.3934	0.3333	31	0.4609
0.3151	0.6667	62	0.4033
0.4147	1.0	93	0.3788
0.3355	1.3333	124	0.3674
0.3675	1.6667	155	0.3620
0.3235	2.0	186	0.3609

Framework versions

• PEFT 0.15.2
• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.5.1
• Tokenizers 0.21.1