adapter: lora
base_model: migtissera/Tess-v2.5-Phi-3-medium-128k-14B
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - c7b2561c26d35f95_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/c7b2561c26d35f95_train_data.json
  type:
    field_instruction: instruction
    field_output: output
    format: '{instruction}'
    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/c95dc644-df33-4332-ac03-7b9ed8c47c1e
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: 0.1
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: 588
micro_batch_size: 4
mlflow_experiment_name: /tmp/c7b2561c26d35f95_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: 0.04267012579153083
wandb_entity: null
wandb_mode: online
wandb_name: efc962aa-703d-4b90-956b-ce0d478c6d93
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: efc962aa-703d-4b90-956b-ce0d478c6d93
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

c95dc644-df33-4332-ac03-7b9ed8c47c1e

This model is a fine-tuned version of migtissera/Tess-v2.5-Phi-3-medium-128k-14B on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.1986

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: 588

Training results

Training Loss	Epoch	Step	Validation Loss
5.1165	0.0003	1	0.6117
1.8033	0.0285	100	0.2168
1.4891	0.0571	200	0.2111
1.9368	0.0856	300	0.2064
1.619	0.1141	400	0.2011
1.4051	0.1426	500	0.1986

Framework versions

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