adapter: lora
base_model: microsoft/Phi-3-mini-4k-instruct
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 117018a141f18637_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/117018a141f18637_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
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 4
flash_attention: false
fp16: false
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: true
hub_model_id: dsakerkwq/95da7394-629e-44d0-a881-33b65930c124
learning_rate: 0.0002
load_in_4bit: false
load_in_8bit: false
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_memory:
  0: 75GiB
max_steps: 30
micro_batch_size: 2
mlflow_experiment_name: /tmp/117018a141f18637_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: adamw_torch
output_dir: miner_id_24
pad_to_sequence_len: true
s2_attention: false
sample_packing: false
saves_per_epoch: 4
sequence_len: 2048
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: 95da7394-629e-44d0-a881-33b65930c124
wandb_project: Gradients-On-Demand
wandb_runid: 95da7394-629e-44d0-a881-33b65930c124
warmup_steps: 100
weight_decay: 0.01
xformers_attention: false

95da7394-629e-44d0-a881-33b65930c124

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: 1.3129

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: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 8
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: 100
training_steps: 30

Training results

Training Loss	Epoch	Step	Validation Loss
9.5851	0.0001	1	1.5901
7.1224	0.0002	3	1.5890
9.5363	0.0004	6	1.5892
7.9852	0.0006	9	1.5890
10.2803	0.0008	12	1.5816
7.6296	0.0010	15	1.5708
7.8796	0.0012	18	1.5497
9.9004	0.0014	21	1.5139
9.1403	0.0015	24	1.4538
10.1342	0.0017	27	1.3818
12.7854	0.0019	30	1.3129

Framework versions

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

1-lock/95da7394-629e-44d0-a881-33b65930c124

95da7394-629e-44d0-a881-33b65930c124

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

1-lock/
95da7394-629e-44d0-a881-33b65930c124