See axolotl config

axolotl version: 0.4.1

adapter: lora
base_model: bigscience/bloomz-560m
bf16: true
chat_template: llama3
datasets:
- data_files:
  - eb75b6ffdc77ea4d_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/eb75b6ffdc77ea4d_train_data.json
  type:
    field_input: input
    field_instruction: instruction
    field_output: output
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 2
eval_batch_size: 2
eval_max_new_tokens: 128
eval_steps: 5
flash_attention: false
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: sn56a1/d4fabac2-e3e1-4fff-a1b4-6257bf0b2766
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 5.0e-05
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 16
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 8
lora_target_linear: true
lr_scheduler: cosine
max_steps: 200
micro_batch_size: 2
mlflow_experiment_name: /tmp/eb75b6ffdc77ea4d_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
optimizer_betas:
- 0.9
- 0.999
optimizer_epsilon: 1e-08
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 10
seed: 3848712367
sequence_len: 512
shuffle: true
strict: false
tf32: false
tokenizer_type: AutoTokenizer
torch_compile: true
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: sn56-miner
wandb_mode: disabled
wandb_name: null
wandb_project: god
wandb_run: trfv
wandb_runid: null
warmup_steps: 5
weight_decay: 0.0
xformers_attention: null

d4fabac2-e3e1-4fff-a1b4-6257bf0b2766

This model is a fine-tuned version of bigscience/bloomz-560m on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.6999

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: 5e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 3848712367
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• total_eval_batch_size: 4
• 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: 5
• training_steps: 200

Training results

Training Loss	Epoch	Step	Validation Loss
7.4453	0.0002	1	2.1528
8.543	0.0008	5	2.1259
8.043	0.0016	10	2.0549
7.668	0.0024	15	1.9907
7.6133	0.0032	20	1.9341
7.7461	0.0040	25	1.8898
7.4922	0.0049	30	1.8510
8.6523	0.0057	35	1.8164
7.5938	0.0065	40	1.7933
6.8398	0.0073	45	1.7806
7.0586	0.0081	50	1.7699
6.5078	0.0089	55	1.7592
6.5625	0.0097	60	1.7515
6.7969	0.0105	65	1.7444
7.5508	0.0113	70	1.7395
6.6484	0.0121	75	1.7368
6.207	0.0130	80	1.7311
6.9844	0.0138	85	1.7265
6.5781	0.0146	90	1.7229
6.7773	0.0154	95	1.7205
6.6094	0.0162	100	1.7176
6.3438	0.0170	105	1.7143
5.8633	0.0178	110	1.7127
6.6172	0.0186	115	1.7118
6.5234	0.0194	120	1.7095
6.8672	0.0202	125	1.7075
6.5039	0.0211	130	1.7050
6.2188	0.0219	135	1.7045
6.4609	0.0227	140	1.7034
6.5938	0.0235	145	1.7017
6.793	0.0243	150	1.7012
6.4512	0.0251	155	1.7010
6.2852	0.0259	160	1.6996
6.793	0.0267	165	1.6995
7.0039	0.0275	170	1.7002
6.9258	0.0283	175	1.6999

Framework versions

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