metadata
library_name: peft
license: llama3.2
base_model: meta-llama/Llama-3.2-3B-Instruct
tags:
- axolotl
- generated_from_trainer
datasets:
- minpeter/xlam-function-calling-60k-hermes
- minpeter/xlam-irrelevance-7.5k-qwen2.5-72b-distill-hermes
model-index:
- name: m-3b-v1-iteration-00-sf-xlam-07
results: []
See axolotl config
axolotl version: 0.7.0
base_model: meta-llama/Llama-3.2-3B-Instruct
hub_model_id: morsmordre/m-3b-v1-iteration-00-sf-xlam-07
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
chat_template: llama3
field_messages: conversations
message_field_role: from
message_field_content: value
shards: 30
- 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: 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: 16
lora_alpha: 32
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: 1
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|>"
m-3b-v1-iteration-00-sf-xlam-07
This model is a fine-tuned version of meta-llama/Llama-3.2-3B-Instruct on the minpeter/xlam-function-calling-60k-hermes and the minpeter/xlam-irrelevance-7.5k-qwen2.5-72b-distill-hermes datasets. It achieves the following results on the evaluation set:
- Loss: 0.2159
Model description
| Test Category | Adapter Accuracy | Base Model Accuracy | Improvement |
|---|---|---|---|
| irrelevance | 76.25% | 72.08% | +4.17% |
| parallel_multiple | 89.50% | 10.00% | +79.50% |
| parallel | 89.50% | 11.50% | +78.00% |
| simple | 92.75% | 24.75% | +68.00% |
| multiple | 93.50% | 20.00% | +73.50% |
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: 1.0
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.8199 | 0.0072 | 1 | 0.6489 |
| 0.0898 | 0.3381 | 47 | 0.2228 |
| 0.2114 | 0.6763 | 94 | 0.2159 |
Framework versions
- PEFT 0.14.0
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0