Hugging Face's logo

l3utterfly
/
phi-2-layla-v1

Text Generation
Transformers
Safetensors
English
phi
custom_code
text-generation-inference
Inference Endpoints
Model card Files Community
1
phi-2-layla-v1 / README.md
l3utterfly's picture
l3utterfly
fixed eos token
30339f8
preview code
|
raw history blame
3.63 kB
metadata
tags:
  - generated_from_trainer
model-index:
  - name: out
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.4.0

base_model: /home/layla/src/text-generation-webui/models/phi-2
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: /home/layla/src/Layla-datasets/datasets_formatted/base/dailydialog.topicalchat.teatime.openhermes.jsonl
    ds_type: json # see other options below
    type: sharegpt
    conversation: vicuna_v1.1

# datasets:
#   - path: /home/layla/src/Layla-datasets/datasets_formatted/airoboros_alpaca.jsonl
#     type: alpaca

dataset_prepared_path: last_run_prepared
val_set_size: 0.01
output_dir: ./out

sequence_len: 2048
sample_packing: true
pad_to_sequence_len: true

wandb_project:
wandb_entity:
wandb_watch:
wandb_run_id:
wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 1
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.0000005

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: true

gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: True
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_ratio: 0.05
eval_steps: 0.1
eval_sample_packing: true
save_steps: 300
debug:
deepspeed: /home/layla/src/Layla-datasets/axolotl/configs/deepspeed/zero2.json # multi-gpu only
weight_decay: 0.0
fsdp:
fsdp_config:
resize_token_embeddings_to_32x: true
special_tokens:
  bos_token: "<|endoftext|>"
  eos_token: "<|endoftext|>"
  unk_token: "<|endoftext|>"
  pad_token: "<|endoftext|>"

out

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.8072

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: 5e-07
  • train_batch_size: 2
  • eval_batch_size: 2
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 5
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 40
  • total_eval_batch_size: 10
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 17
  • num_epochs: 1

Training results

Training Loss Epoch Step Validation Loss
0.9616 0.0 1 1.0031
0.9489 0.1 372 0.8825
0.987 0.2 744 0.8487
0.818 0.3 1116 0.8313
0.8389 0.4 1488 0.8212
0.9015 0.5 1860 0.8146
0.8237 0.6 2232 0.8108
0.7562 0.7 2604 0.8088
0.8776 0.8 2976 0.8078
0.8703 0.9 3348 0.8072

Framework versions

  • Transformers 4.39.0.dev0
  • Pytorch 2.2.0
  • Datasets 2.17.1
  • Tokenizers 0.15.0