metadata
library_name: peft
tags:
- generated_from_trainer
base_model: ./lora-logo_real_filter_line_12_ds7b_ds33i_lr_0.0002_alpha_512_r_512_merged
model-index:
- name: >-
lora-logo_adapt_real_fix_continue_filter_line_12_ds7b_ds33i_lr_0.0002_alpha_512_r_512
results: []
See axolotl config
axolotl version: 0.4.0
adapter: lora
base_model: ./lora-logo_real_filter_line_12_ds7b_ds33i_lr_0.0002_alpha_512_r_512_merged
bf16: auto
dataset_prepared_path: ./logo_ds_preprocess_list_gpt35
datasets:
- path: ../logo/adapt_deepseek_filter_line_12_synthetic_training_data_32k.jsonl
type:
field_instruction: input
field_output: output
format: '### Instruction:
{input}
### Response:
'
no_input_format: '{instruction}'
debug: null
deepspeed: ./deepspeed_configs/zero2.json
early_stopping_patience: null
eval_sample_packing: true
evals_per_epoch: 4
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
group_by_length: false
is_llama_derived_model: true
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: true
local_rank: null
logging_steps: 1
lora_alpha: 512
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 512
lora_target_linear: true
lr_scheduler: cosine
micro_batch_size: 8
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_bnb_8bit
output_dir: ./lora-logo_adapt_real_fix_continue_filter_line_12_ds7b_ds33i_lr_0.0002_alpha_512_r_512
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: true
saves_per_epoch: 1
sequence_len: 1800
special_tokens:
bos_token: "<\uFF5Cbegin\u2581of\u2581sentence\uFF5C>"
eos_token: <|EOT|>
strict: true
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
val_set_size: 0.05
wandb_entity: null
wandb_log_model: null
wandb_name: logo_adapt_real_fix_continue_filter_line_12_ds7b_ds33i_lr_0.0002_alpha_512_r_512
wandb_project: pbe-axo
wandb_watch: null
warmup_steps: 20
weight_decay: 0.0
xformers_attention: null
lora-logo_adapt_real_fix_continue_filter_line_12_ds7b_ds33i_lr_0.0002_alpha_512_r_512
This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4632
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: 8
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 64
- total_eval_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 20
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.4154 | 0.0 | 1 | 0.4359 |
| 0.4055 | 0.25 | 78 | 0.4264 |
| 0.4281 | 0.5 | 156 | 0.4272 |
| 0.3995 | 0.75 | 234 | 0.4240 |
| 0.3828 | 1.01 | 312 | 0.4218 |
| 0.3811 | 1.24 | 390 | 0.4272 |
| 0.3738 | 1.49 | 468 | 0.4268 |
| 0.3538 | 1.74 | 546 | 0.4242 |
| 0.3657 | 1.99 | 624 | 0.4205 |
| 0.287 | 2.22 | 702 | 0.4607 |
| 0.2472 | 2.47 | 780 | 0.4616 |
| 0.2541 | 2.72 | 858 | 0.4632 |
Framework versions
- PEFT 0.10.0
- Transformers 4.40.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.15.0
- Tokenizers 0.15.0