You can deploy this model and start earning money today!

Alphatao/
1200b65c-885d-4242-9990-175b9dc6e085

Name: 1200b65c-885d-4242-9990-175b9dc6e085
Availability: InStock

generated_from_trainer

JackFram/llama-160m

adapter

apache-2.0

Views

No views yet

See axolotl config

axolotl version: 0.4.1

yaml

adapter: lora
base_model: JackFram/llama-160m
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - 08b9077a9997bb5d_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/08b9077a9997bb5d_train_data.json
  type:
    field_instruction: question
    field_output: answer
    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: true
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: Alphatao/1200b65c-885d-4242-9990-175b9dc6e085
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: 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
- o_proj
- down_proj
- up_proj
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 3600
micro_batch_size: 4
mlflow_experiment_name: /tmp/08b9077a9997bb5d_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: </s>
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: 63c93655-01ce-4a07-98c8-c019b4e4e4b2
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 63c93655-01ce-4a07-98c8-c019b4e4e4b2
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

1200b65c-885d-4242-9990-175b9dc6e085

This model is a fine-tuned version of JackFram/llama-160m on the None dataset. It achieves the following results on the evaluation set:

Loss: 2.5791

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: 1690

Training results

Training Loss	Epoch	Step	Validation Loss
3.4779	0.0012	1	3.8497
2.6745	0.1184	100	2.7417
2.3918	0.2367	200	2.6902
2.442	0.3551	300	2.6656
2.6763	0.4734	400	2.6464
2.5149	0.5918	500	2.6323
2.6993	0.7102	600	2.6184
2.8935	0.8285	700	2.6073
2.3858	0.9469	800	2.5994
2.3872	1.0658	900	2.5960
2.4632	1.1842	1000	2.5913
2.4971	1.3026	1100	2.5877
2.2192	1.4209	1200	2.5850
2.5662	1.5393	1300	2.5823
2.4268	1.6576	1400	2.5805
2.4365	1.7760	1500	2.5793
2.1672	1.8944	1600	2.5791

Framework versions

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