You can deploy this model and start earning money today!

Alphatao/
600af72b-4398-4531-9403-d5a23539543a

Name: 600af72b-4398-4531-9403-d5a23539543a
Availability: InStock

generated_from_trainer

elyza/Llama-3-ELYZA-JP-8B

adapter

llama3

Views

No views yet

See axolotl config

axolotl version: 0.4.1

yaml

adapter: lora
base_model: elyza/Llama-3-ELYZA-JP-8B
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 5e3277732f6641a2_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/5e3277732f6641a2_train_data.json
  type:
    field_input: keyword
    field_instruction: problem
    field_output: outputs
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: Alphatao/600af72b-4398-4531-9403-d5a23539543a
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_best_model_at_end: true
load_in_4bit: false
load_in_8bit: false
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
lora_target_modules:
- q_proj
- k_proj
- v_proj
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 840
micro_batch_size: 4
mlflow_experiment_name: /tmp/5e3277732f6641a2_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 2
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 100
sequence_len: 2048
special_tokens:
  pad_token: <|eot_id|>
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.03631082062454612
wandb_entity: null
wandb_mode: online
wandb_name: 5003b58d-7b70-42b3-aa07-ea56bbef0106
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 5003b58d-7b70-42b3-aa07-ea56bbef0106
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

600af72b-4398-4531-9403-d5a23539543a

This model is a fine-tuned version of elyza/Llama-3-ELYZA-JP-8B on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.5972

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: 0.0002
train_batch_size: 4
eval_batch_size: 4
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_BNB 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: 10
training_steps: 840

Training results

Training Loss	Epoch	Step	Validation Loss
1.1481	0.0002	1	1.1430
0.7187	0.0241	100	0.7332
0.6863	0.0482	200	0.6811
0.6909	0.0723	300	0.6514
0.531	0.0965	400	0.6297
0.558	0.1206	500	0.6153
0.6073	0.1447	600	0.6049
0.6223	0.1688	700	0.5990
0.5716	0.1929	800	0.5972

Framework versions

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