shisa-v1-phi3-14b / README.md
leonardlin's picture
Update README.md
3b47a24 verified
metadata
license: mit
base_model: microsoft/Phi-3-medium-128k-instruct
tags:
  - generated_from_trainer
model-index:
  - name: outputs/phi3-medium-128k-14b.8e6
    results: []

This test ablation probably shouldn't be used. It actually underperforms the original Phi 3 Medium Instruct model as it was trained w/ ChatML (but still performs better with the original Phi 3 instruct template). If you were to do a training run w/ this dataset, I'd recommend modifying the training to use the Phi 3 Instruct chat format.

Performance

Measured using a fork of Lightblue's Shaberi benchmark framework:

Model Average ELYZA-tasks-100 MT-Bench Rakuda Tengu-Bench
gpt-4-turbo-2024-04-09 8.75 8.78 8.74 9.18 8.31
gpt-4o-2024-05-13 8.72 8.88 8.69 9.15 8.16
gemini-1.5-pro 8.58 8.58 8.93 9.20 7.61
claude-3-opus-20240229 8.55 8.64 8.58 8.75 8.23
CohereForAI/c4ai-command-r-plus 7.69 7.50 7.43 9.05 6.79
shisa-ai/shisa-v1-llama3-70b 7.30 7.34 7.67 8.15 6.04
gpt-3.5-turbo-0125 7.17 7.24 6.98 7.64 6.82
shisa-ai/shisa-v1-llama3-70b.2e5 7.17 7.16 7.45 7.98 6.09
karakuri-ai/karakuri-lm-8x7b-chat-v0.1 7.00 7.18 6.30 7.98 6.55
karakuri-ai/karakuri-lm-70b-chat-v0.1 6.84 6.86 6.43 7.85 6.23
lightblue/ao-karasu-72B 6.81 7.19 6.54 7.25 6.27
shisa-ai/shisa-v1-llama3-8b 6.59 6.67 6.95 7.05 5.68
microsoft/Phi-3-medium-128k-instruct 6.48 7.10 5.92 6.84 6.04
shisa-ai/shisa-swallowmx-13a47b-v1 6.17 6.48 6.07 7.11 5.03
lightblue/suzume-llama-3-8B-japanese 5.96 6.68 4.96 6.68 5.53
augmxnt/shisa-gamma-7b-v1 5.82 5.96 5.02 6.85 5.47
shisa-ai/shisa-v1-phi3-14b 5.77 6.28 5.26 6.55 5.01
shisa-ai/shisa-v1-gemma-8b 5.64 6.50 5.42 5.10 5.55
Rakuten/RakutenAI-7B-chat 5.58 5.92 4.60 6.58 5.24
lightblue/qarasu-14B-chat-plus-unleashed 5.20 5.58 4.74 5.46 5.01
shisa-ai/shisa-v1-mistral0.3-7b 5.11 5.64 6.10 3.83 4.86
cyberagent/calm2-7b-chat 4.76 4.90 3.58 5.75 4.81
mistralai/Mistral-7B-Instruct-v0.2 4.69 5.78 4.65 3.80 4.53
shisa-ai/shisa-v1-yi1.5-9b 4.63 5.98 4.28 3.26 5.00
augmxnt/shisa-7b-v1 4.50 4.63 3.95 4.89 4.53

Built with Axolotl

See axolotl config

axolotl version: 0.4.0

base_model: microsoft/Phi-3-medium-128k-instruct
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
trust_remote_code: true

load_in_8bit: false
load_in_4bit: false
strict: false

use_wandb: true
wandb_project: shisa-v2
wandb_entity: augmxnt
wandb_name: shisa-llama3-70b-v1.8e6

chat_template: chatml
datasets:
  - path: augmxnt/ultra-orca-boros-en-ja-v1
    type: sharegpt
dataset_prepared_path: last_run_prepared
val_set_size: 0.05
output_dir: ./outputs/phi3-medium-128k-14b.8e6

sequence_len: 4096
sample_packing: true
pad_to_sequence_len: true
neftune_noise_alpha: 5

gradient_accumulation_steps: 4
micro_batch_size: 2
num_epochs: 3
optimizer: paged_adamw_8bit
adam_beta2: 0.95
adam_epsilon: 0.00001
max_grad_norm: 1.0
lr_scheduler: linear
learning_rate: 0.000008

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_steps: 100
evals_per_epoch: 4
saves_per_epoch: 1
debug:
deepspeed: axolotl/deepspeed_configs/zero3_bf16.json
weight_decay: 0.1
fsdp:
fsdp_config:
resize_token_embeddings_to_32x: true
special_tokens:
  pad_token: "<|endoftext|>"

outputs/phi3-medium-128k-14b.8e6

This model is a fine-tuned version of microsoft/Phi-3-medium-128k-instruct on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3339

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: 8e-06
  • train_batch_size: 2
  • eval_batch_size: 2
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 8
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 64
  • total_eval_batch_size: 16
  • optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-05
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 100
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
2.8309 0.0021 1 2.3406
0.7688 0.2513 121 0.4958
0.6435 0.5026 242 0.3830
0.5286 0.7539 363 0.3626
0.5559 1.0052 484 0.3549
0.4651 1.2425 605 0.3486
0.5294 1.4938 726 0.3432
0.5453 1.7451 847 0.3392
0.5258 1.9964 968 0.3376
0.4805 2.2331 1089 0.3357
0.4552 2.4844 1210 0.3352
0.5358 2.7357 1331 0.3339

Framework versions

  • Transformers 4.40.2
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1