See axolotl config

axolotl version: 0.4.1

adam_beta2: 0.95
adam_epsilon: 1.0e-05
adapter: qlora
base_model: microsoft/phi-1_5
dataset_prepared_path: null
datasets:
- path: garage-bAInd/Open-Platypus
  type: alpaca
debug: null
deepspeed: null
early_stopping_patience: null
evals_per_epoch: 1
flash_attention: true
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: true
hub_model_id: AdamRTomkins/test_upload
hub_strategy: end
learning_rate: 3.0e-06
load_in_4bit: true
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 2
micro_batch_size: 1
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_torch
output_dir: ./outputs/phi-sft-out
pad_to_sequence_len: true
resize_token_embeddings_to_32x: true
resume_from_checkpoint: null
sample_packing: true
saves_per_epoch: 1
sequence_len: 1024
special_tokens:
  pad_token: <|endoftext|>
strict: false
tokenizer_type: AutoTokenizer
val_set_size: 0.05
wandb_entity: null
wandb_log_model: null
wandb_name: null
wandb_project: null
wandb_watch: null
warmup_steps: 100
weight_decay: 0.1
xformers_attention: null

test_upload

This model is a fine-tuned version of microsoft/phi-1_5 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.3469

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: 3e-06
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-05
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• training_steps: 2

Training results

Training Loss	Epoch	Step	Validation Loss
1.6676	0.0002	2	1.3469

Framework versions

• PEFT 0.11.1
• Transformers 4.41.1
• Pytorch 2.1.2+cu118
• Datasets 2.19.1
• Tokenizers 0.19.1