adapter: lora
base_model: microsoft/Phi-3-mini-4k-instruct
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 2470316c9e2e34b8_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/2470316c9e2e34b8_train_data.json
  type:
    field_input: src_lang
    field_instruction: src_sent
    field_output: tgt_sent
    format: '{instruction} {input}'
    no_input_format: '{instruction}'
    system_format: '{system}'
    system_prompt: ''
debug: null
device_map:
  ? ''
  : 0,1,2,3,4,5,6,7
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
flash_attention: false
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: Alphatao/2081438a-078a-42bf-980c-2dd8069065f5
hub_repo: null
hub_strategy: null
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: 128
lora_dropout: 0.3
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 64
lora_target_linear: true
lora_target_modules:
- q_proj
- k_proj
- v_proj
- o_proj
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 1734
micro_batch_size: 4
mlflow_experiment_name: /tmp/2470316c9e2e34b8_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: 1024
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.016700066800267203
wandb_entity: null
wandb_mode: online
wandb_name: 40b5f6dd-bdee-45dd-9390-5a59b7a4a161
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 40b5f6dd-bdee-45dd-9390-5a59b7a4a161
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

2081438a-078a-42bf-980c-2dd8069065f5

This model is a fine-tuned version of microsoft/Phi-3-mini-4k-instruct on the None dataset. It achieves the following results on the evaluation set:

Loss: 1.0834

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: 1734

Training results

Training Loss	Epoch	Step	Validation Loss
27.6398	0.0001	1	3.4388
16.3097	0.0109	100	1.8248
10.6714	0.0217	200	1.6117
13.508	0.0326	300	1.5078
11.1271	0.0435	400	1.4314
11.2775	0.0543	500	1.3530
8.6485	0.0652	600	1.3029
11.3778	0.0761	700	1.2563
11.0668	0.0870	800	1.2160
10.1429	0.0978	900	1.1866
10.7408	0.1087	1000	1.1618
7.8018	0.1196	1100	1.1419
8.9112	0.1304	1200	1.1201
8.0834	0.1413	1300	1.1052
9.3992	0.1522	1400	1.0941
10.2637	0.1630	1500	1.0878
9.8593	0.1739	1600	1.0844
8.3447	0.1848	1700	1.0834

Framework versions

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