See axolotl config
axolotl version: 0.4.0
base_model: openbmb/MiniCPM-2B-sft-bf16
load_in_8bit: false
load_in_4bit: false
strict: false
push_dataset_to_hub:
datasets:
- path: teknium/GPT4-LLM-Cleaned
type: alpaca
dataset_prepared_path:
val_set_size: 0.05
adapter:
lora_model_dir:
sequence_len: 4096
max_packed_sequence_len:
lora_r: 8
lora_alpha: 32
lora_dropout: 0.05
lora_target_modules:
lora_target_linear: true
lora_fan_in_fan_out:
wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:
output_dir: ./qlora-out
gradient_accumulation_steps: 2
micro_batch_size: 2
num_epochs: 1.5
optimizer: paged_adamw_8bit
torchdistx_path:
lr_scheduler: cosine
learning_rate: 0.0001
train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: true
gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention: true
flash_attention:
gptq_groupsize:
gptq_model_v1:
warmup_steps: 10
evals_per_epoch: 2
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.1
fsdp:
fsdp_config:
special_tokens:
trust_remote_code: true
qlora-out
This model is a fine-tuned version of openbmb/MiniCPM-2B-sft-bf16 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0525
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: 0.0001
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- total_eval_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: 1.5
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.0903 | 0.0 | 1 | 1.7199 |
| 0.8959 | 0.5 | 1620 | 1.1007 |
| 0.995 | 1.0 | 3240 | 1.0342 |
| 0.864 | 1.5 | 4860 | 1.0525 |
Framework versions
- Transformers 4.38.0.dev0
- Pytorch 2.1.2+cu118
- Datasets 2.16.1
- Tokenizers 0.15.0
- Downloads last month
- 32
Inference API (serverless) does not yet support model repos that contain custom code.
Model tree for Crystalcareai/CrystalMiniCPM
Base model
openbmb/MiniCPM-2B-sft-bf16