See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: 01-ai/Yi-1.5-9B-Chat-16K
bf16: true
chat_template: llama3
datasets:
- data_files:
  - dbe8600da51bffa0_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/dbe8600da51bffa0_train_data.json
  type:
    field_instruction: premise_en
    field_output: hypothesis_en
    format: '{instruction}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 5
eval_table_size: null
flash_attention: false
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: false
group_by_length: false
hub_model_id: lesso10/15d67be8-501f-4311-ad4d-20e3d686980b
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: true
local_rank: null
logging_steps: 1
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lr_scheduler: cosine
max_steps: 50
micro_batch_size: 4
mlflow_experiment_name: /tmp/dbe8600da51bffa0_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_torch
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 10
sequence_len: 1024
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 5e32b8d2-c60e-41d9-b8a7-7848aed67042
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 5e32b8d2-c60e-41d9-b8a7-7848aed67042
warmup_steps: 5
weight_decay: 0.0
xformers_attention: null

15d67be8-501f-4311-ad4d-20e3d686980b

This model is a fine-tuned version of 01-ai/Yi-1.5-9B-Chat-16K on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.3578

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Use OptimizerNames.ADAMW_TORCH 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: 5
• training_steps: 50

Training results

Training Loss	Epoch	Step	Validation Loss
4.5589	0.0018	1	4.2694
3.809	0.0091	5	3.4923
1.6845	0.0182	10	1.7621
1.477	0.0274	15	1.5267
1.3267	0.0365	20	1.4556
1.3047	0.0456	25	1.4177
1.4743	0.0547	30	1.3929
1.3607	0.0639	35	1.3760
0.9065	0.0730	40	1.3685
1.3038	0.0821	45	1.3589
1.4639	0.0912	50	1.3578

Framework versions

• PEFT 0.13.2
• Transformers 4.46.0
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1