metadata
license: mit
base_model: microsoft/phi-2
tags:
- axolotl
- generated_from_trainer
model-index:
- name: evol-codealpaca-pairwise-sharegpt-test
results: []
See axolotl config
axolotl version: 0.3.0
base_model: microsoft/phi-2
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
trust_remote_code: true
hub_model_id: AlekseyKorshuk/evol-codealpaca-pairwise-sharegpt-test
load_in_8bit: false
load_in_4bit: false
strict: false
datasets:
- path: AlekseyKorshuk/evol-codealpaca-pairwise-sharegpt
type: sharegpt
conversation: chatml
dataset_prepared_path:
val_set_size: 0.001
output_dir: ./phi-sft-out
sequence_len: 2048
sample_packing: false # currently unsupported
pad_to_sequence_len:
lora_r:
lora_alpha:
lora_dropout:
lora_target_modules:
lora_target_linear:
lora_fan_in_fan_out:
wandb_project: ui-thesis
wandb_entity:
wandb_watch:
wandb_name: phi-2-chatml-test
wandb_log_model:
gradient_accumulation_steps: 1
micro_batch_size: 16
num_epochs: 3
optimizer: paged_adamw_8bit
adam_beta1: 0.9
adam_beta2: 0.95
adam_epsilon: 0.00001
#max_grad_norm: 1.0
lr_scheduler: cosine
learning_rate: 1e-5
warmup_ratio: 0.03
weight_decay: 0.01
train_on_inputs: false
group_by_length: false
bf16: false
fp16: false
tf32: false
float16: true
gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true
evals_per_epoch: 1
eval_table_size: 8 # Approximate number of predictions sent to wandb depending on batch size. Enabled above 0. Default is 0
eval_table_max_new_tokens: 512 # Total number of tokens generated for predictions sent to wandb. Default is 128
saves_per_epoch: 1
save_total_limit: 1
debug:
deepspeed:
fsdp:
fsdp_config:
resize_token_embeddings_to_32x: true
special_tokens:
eos_token: "<|im_end|>"
pad_token: "<|endoftext|>"
tokens:
- "<|im_start|>"
evol-codealpaca-pairwise-sharegpt-test
This model is a fine-tuned version of microsoft/phi-2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0374
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- total_train_batch_size: 128
- total_eval_batch_size: 128
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-05
- lr_scheduler_type: cosine
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.0571 | 0.01 | 1 | 1.2056 |
| 0.8271 | 1.0 | 82 | 1.0443 |
| 0.7871 | 2.0 | 164 | 1.0378 |
| 0.8198 | 3.0 | 246 | 1.0374 |
Framework versions
- Transformers 4.37.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.16.1
- Tokenizers 0.15.0