Open-Assistant稳定LM-7B SFT-7模型

这是 Open-Assistant 项目的第7次迭代的英文监督细调（SFT）模型。它基于一个在2023年4月12日之前通过 https://open-assistant.io/ 人类反馈Web应用程序收集的助手对话的人类示范的StableLM 7B进行了微调。

模型详情

• 开发者： Open-Assistant Contributors
• 模型类型：基于Transformer的语言模型
• 语言：英语
• 微调自： stabilityai/stablelm-base-alpha-7b
• 代码： Open-Assistant/model/model_training
• 演示：TODO
• 许可证：知识共享许可证（ CC BY-SA-4.0 ）
• 联系方式： Open-Assistant Discord

提示

有两个特殊令牌用于标记用户和助手交替的开始：<|prompter|>和<|assistant|>。每个交替以<|endoftext|>令牌结束。

输入提示示例：

<|prompter|>What is a meme, and what's the history behind this word?<|endoftext|><|assistant|>

输入以<|assistant|>令牌结尾，以表示模型应开始生成助手的回复。

开发详情

• wandb： https://wandb.ai/open-assistant/supervised-finetuning/runs/08dfhyuc
• 基础模型： stabilityai/stablelm-base-alpha-7b
• 检查点：3个时代（12000步）

命令：deepspeed trainer_sft.py --configs defaults stablelm-7b oasst-mix --cache_dir /home/ubuntu/data_cache --output_dir .saved/stable-lm-7b-1 --num_train_epochs 4 --deepspeed

数据：

oasst-mix:
  save_strategy: epoch
  sort_by_length: false
  use_custom_sampler: false
  datasets:
    - oasst_export:
        lang: "bg,ca,cs,da,de,en,es,fr,hr,hu,it,nl,pl,pt,ro,ru,sl,sr,sv,uk"
        input_file_path: 2023-04-12_oasst_release_ready_synth.jsonl.gz
    - vicuna:
        val_split: 0.05
        max_val_set: 800
        fraction: 1.0
    - dolly15k:
        val_split: 0.05
        max_val_set: 300
    - grade_school_math_instructions:
        val_split: 0.05
    - code_alpaca:
        val_split: 0.05
        max_val_set: 250

stablelm：

stablelm-7b:
  dtype: fp16
  log_dir: stablelm_log_7b
  model_name: stabilityai/stablelm-base-alpha-7b
  output_dir: stablelm_7b
  max_length: 4096
  warmup_steps: 100
  gradient_checkpointing: true
  gradient_accumulation_steps: 2
  per_device_train_batch_size: 4
  per_device_eval_batch_size: 4
  eval_steps: 100
  save_steps: 500
  num_train_epochs: 4
  save_total_limit: 4
  use_flash_attention: true

零配置：

{
  "fp16": {
    "enabled": "auto",
    "loss_scale": 0,
    "loss_scale_window": 1000,
    "initial_scale_power": 16,
    "hysteresis": 2,
    "min_loss_scale": 1
  },
  "bf16": {
    "enabled": "auto"
  },
  "optimizer": {
    "type": "AdamW",
    "params": {
      "lr": "auto",
      "betas": "auto",
      "eps": "auto",
      "weight_decay": "auto"
    }
  },
  "scheduler": {
    "type": "WarmupDecayLR",
    "params": {
      "warmup_min_lr": "auto",
      "warmup_max_lr": "auto",
      "warmup_num_steps": "auto",
      "total_num_steps": "auto"
    }
  },
  "zero_optimization": {
    "stage": 2,
    "allgather_partitions": true,
    "allgather_bucket_size": 1e9,
    "overlap_comm": false,
    "reduce_scatter": true,
    "reduce_bucket_size": 1e9,
    "contiguous_gradients": true
  },
  "gradient_accumulation_steps": "auto",
  "gradient_clipping": "auto",
  "steps_per_print": 2000,
  "train_batch_size": "auto",
  "train_micro_batch_size_per_gpu": "auto",
  "wall_clock_breakdown": false
}