base_model: "/workspace/data/pyx_14_03_02/checkpoint-5738"
model_type: Qwen2ForCausalLM
tokenizer_type: AutoTokenizer

#plugins:
#  - axolotl.integrations.liger.LigerPlugin

load_in_8bit: false
load_in_4bit: false
strict: false

chat_template: chatml

datasets:
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/think_6061.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value

  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/argila_dpo_chat_message.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value

  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/dpo_domen_all.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/dpo_dostoevskiy.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/forum.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/forum_2.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/forum_3.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/mes_1468.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/no_yes.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/done_narkolog.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
      
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/prompt_5152.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/role_1292.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/system.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/answer_only_moder.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    
  - path: "/workspace/data/axolotl/examples/qwen2/lora"
    data_files: "./lora/context_515.jsonl"
    type: chat_template
    field_messages: conversations
    message_field_role: from
    message_field_content: value
    

dataset_prepared_path: "/workspace/data/axolotl/examples/qwen2"
val_set_size: 0.05
output_dir: "/workspace/data/pyx_14_05_02"

sequence_len: 3400
sample_packing: false
pad_to_sequence_len: true



wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 8
micro_batch_size: 2
num_epochs: 1

optimizer: paged_adamw_8bit
lr_scheduler: cosine
cosine_min_lr_ratio: 1.0
cosine_constant_lr_ratio: 1.0
learning_rate: 0.00002

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false
max_grad_norm: 2

gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
logging_steps: 1
xformers_attention:
flash_attention: false

save_safetensors: true

warmup_steps: 100
evals_per_epoch: 4
eval_table_size:
saves_per_epoch: 1
debug:
deepspeed: #deepspeed_configs/zero3_bf16_cpuoffload_all.json
weight_decay: 0.0
fsdp:
#  - full_shard
#  - auto_wrap
fsdp_config:
#  fsdp_limit_all_gathers: true
#  fsdp_sync_module_states: true
#  fsdp_offload_params: true
#  fsdp_use_orig_params: false
#  fsdp_cpu_ram_efficient_loading: true
#  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
#  fsdp_transformer_layer_cls_to_wrap: LlamaDecoderLayer
#  fsdp_state_dict_type: FULL_STATE_DICT
#  fsdp_sharding_strategy: FULL_SHARD
#  fsdp_backward_prefetch: BACKWARD_PRE
special_tokens:

workspace/data/pyx_14_05_02

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

Loss: 0.8655

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: 2e-05
train_batch_size: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 16
optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 100
num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.6814	0.0002	1	0.8907
1.1736	0.2501	1435	0.9169
0.4295	0.5002	2870	0.8868
0.6204	0.7503	4305	0.8655

Framework versions

Transformers 4.48.1
Pytorch 2.5.1+cu124
Datasets 3.2.0
Tokenizers 0.21.0

Workermen/qwen_14_05_02

workspace/data/pyx_14_05_02

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Workermen/
qwen_14_05_02