phi-smol

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>

<details><summary>See axolotl config</summary>

axolotl version: 0.3.0

base_model: phi-600M-cont/checkpoint-5000
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
trust_remote_code: true

load_in_8bit: false
load_in_4bit: false
strict: false

# max_steps: 8000
#pretraining_dataset: nampdn-ai/tiny-strange-textbooks 
datasets:
  - path: math-ai/StackMathQA
    name: stackmathqa100k
    type:
      system_prompt: ""
      field_system: system
      field_instruction: Q
      field_output: A
      format: "[INST] {instruction} [/INST]"
      no_input_format: "[INST] {instruction} [/INST]"
    train_on_split: train[:10%]
  - path: SciPhi/textbooks-are-all-you-need-lite
    type: completion
    field: completion
    train_on_split: train[:10%]


dataset_prepared_path:
val_set_size: 0.001
output_dir: ./phi-600M-mix

sequence_len: 2048
sample_packing: true  # currently unsupported
pad_to_sequence_len:

adapter:
lora_model_dir:
lora_r:
lora_alpha:
lora_dropout:
lora_target_linear:
lora_fan_in_fan_out:
lora_modules_to_save:

wandb_project: phine
wandb_entity: willfulbytes
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 1
num_epochs: 1
optimizer: paged_adamw_8bit
adam_beta2: 0.98
adam_epsilon: 0.0000001
max_grad_norm: 1.0
lr_scheduler: cosine
learning_rate: 1e-4
cosine_min_lr_ratio: 0.2

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

gradient_checkpointing: true
early_stopping_patience: false
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_steps: 0
evals_per_epoch: 100
saves_per_epoch: 10
save_steps:
debug:
deepspeed:
weight_decay: 0.1
fsdp:
fsdp_config:
resize_token_embeddings_to_32x: true
special_tokens:
  pad_token: "<|endoftext|>"

</details><br>

phi-600M-mix

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

• Loss: 1.6549

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: 0.0001
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-07
• lr_scheduler_type: cosine
• num_epochs: 1

Training results

Framework versions

• Transformers 4.37.0.dev0
• Pytorch 2.0.1
• Datasets 2.16.1
• Tokenizers 0.15.0