Hugging Face's logo

cognitivecomputations
/
dolphin-2.9.3-qwen2-0.5b

Text Generation
Transformers
Safetensors
qwen2
Generated from Trainer
axolotl
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
dolphin-2.9.3-qwen2-0.5b / README.md
ehartford's picture
ehartford
Upload folder using huggingface_hub
4b21382 verified
preview code
|
raw
history blame
4.89 kB
metadata
license: apache-2.0
base_model: Qwen/Qwen2-0.5B
tags:
  - generated_from_trainer
model-index:
  - name: 0.5b out
    results: []

Built with Axolotl

See axolotl config

axolotl version: 0.4.1

base_model: Qwen/Qwen2-0.5B
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

# load_in_4bit: true

chat_template: chatml
datasets:
  - path: /workspace/datasets/dolphin201-sharegpt2.jsonl
    type: sharegpt
    conversation: chatml
  - path: /workspace/datasets/SystemChat_sharegpt.jsonl
    type: sharegpt
    conversation: chatml
  # - path: /workspace/datasets/SystemChat_multilingual_sharegpt.jsonl
  #   type: sharegpt
  #   conversation: chatml
  # - path: /workspace/datasets/SystemChat-2.0-Arabic/SystemChatArabic_sharegpt.jsonl
  #   type: sharegpt
  #   conversation: chatml
  # - path: /workspace/datasets/dolphin-coder-translate-sharegpt2.jsonl
  #   type: sharegpt
  #   conversation: chatml
  # - path: /workspace/datasets/dolphin-coder-codegen-sharegpt2.jsonl
  #   type: sharegpt
  #   conversation: chatml
  # - path: /workspace/datasets/m-a-p_Code-Feedback-sharegpt-unfiltered.jsonl
  #   type: sharegpt
  #   conversation: chatml
  # - path: /workspace/datasets/m-a-p_CodeFeedback-Filtered-Instruction-sharegpt-unfiltered.jsonl
  #   type: sharegpt
    # conversation: chatml
  - path: /workspace/datasets/not_samantha_norefusals.jsonl
    type: sharegpt
    conversation: chatml
  - path: /workspace/datasets/Orca-Math-resort-unfiltered.jsonl
    type: sharegpt
  #   conversation: chatml
  # - path: /workspace/datasets/agent_instruct_react_unfiltered.jsonl
  #   type: sharegpt  
  #   conversation: chatml
  # - path: /workspace/datasets/toolbench_instruct_j1s1_3k_unfiltered.jsonl
  #   type: sharegpt  
  #   conversation: chatml
  # - path: /workspace/datasets/toolbench_negative_unfiltered.jsonl
  #   type: sharegpt
  #   conversation: chatml
  # - path: /workspace/datasets/toolbench_react_10p_unfiltered.jsonl
  #   type: sharegpt
  #   conversation: chatml
  # - path: /workspace/datasets/toolbench_tflan_cot_30p_unfiltered.jsonl
  #   type: sharegpt
  #   conversation: chatml
  - path: /workspace/datasets/openhermes200k_unfiltered.jsonl
    type: sharegpt 
    conversation: chatml

dataset_prepared_path: last_run_prepared
val_set_size: 0.03
output_dir: 0.5b out

sequence_len: 16384
sample_packing: true
pad_to_sequence_len: true

# adapter: qlora
# lora_r: 16
# lora_alpha: 32
# lora_dropout: 0.05
# lora_target_modules:
#   - q_proj
#   - k_proj
#   - v_proj
#   - o_proj
#   - gate_proj
#   - up_proj
#   - down_proj

wandb_project: 2.9.3-qwen-2.9.3-qwen2-0.5b
# wandb_entity: oaaic
# wandb_watch:
# wandb_name:
# wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 1
num_epochs: 3
optimizer: adamw_8bit
lr_scheduler: constant
learning_rate: 1e-4
# max_grad_norm: 1.0

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

gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: true
logging_steps: 1
flash_attention: true
deepspeed: /workspace/axolotl/deepspeed_configs/zero3_bf16.json
warmup_steps: 10
# evals_per_epoch: 2
saves_per_epoch: 2
save_total_limit: 2
weight_decay: 0.1
special_tokens:
  eos_token: <|im_end|>

0.5b out

This model is a fine-tuned version of Qwen/Qwen2-0.5B on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.9948

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
  • distributed_type: multi-GPU
  • num_devices: 8
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • total_eval_batch_size: 8
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: constant
  • lr_scheduler_warmup_steps: 10
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
1.0145 1.0111 933 1.0237
0.9716 2.0116 1867 0.9972
0.8939 2.9743 2766 0.9948

Framework versions

  • Transformers 4.41.1
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1