See axolotl config

axolotl version: 0.4.0

base_model: deepseek-ai/deepseek-coder-6.7b-instruct
model_type: LlamaForCausalLM
tokenizer_type: LlamaTokenizerFast

load_in_8bit: true
load_in_4bit: false
strict: false

datasets:
  - path: vdaita/editpackft_inst_ellipsis
    split: train
    type: oasst
dataset_prepared_path:

test_datasets:
  - path: vdaita/editpackft_inst_ellipsis
    split: test
    type: oasst

output_dir: ./outputs/dscoder-code-ellipsis

sequence_len: 4096
sample_packing: true
pad_to_sequence_len: true

eval_sample_packing: false

adapter: lora
lora_model_dir:
lora_r: 32
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:

lora_modules_to_save:
  - embed_tokens
  - lm_head

wandb_project: huggingface
wandb_log_model: axolotl-dscoder-ellipsis

hub_model_id: vdaita/diff-deepseek-ellipsis
hub_strategy: every_save

gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 1
optimizer: adamw_bnb_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
s2_attention:

warmup_steps: 10
evals_per_epoch: 4
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:

special_tokens:
  bos_token: "<|begin_of_sentence|>"
  eos_token: "<|end_of_sentence|>"
  pad_token: "<|end_of_sentence|>"

diff-deepseek-ellipsis

This model is a fine-tuned version of deepseek-ai/deepseek-coder-6.7b-instruct on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1634

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
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• total_eval_batch_size: 4
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 10
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
0.3241	0.02	1	0.3550
0.2785	0.25	11	0.2303
0.2129	0.51	22	0.1771
0.1803	0.76	33	0.1634

Framework versions

• PEFT 0.10.0
• Transformers 4.40.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.15.0