metadata
library_name: peft
license: apache-2.0
base_model: JackFram/llama-160m
tags:
- axolotl
- generated_from_trainer
model-index:
- name: a5398e13-1cfc-5486-27db-03400d5e39af
results: []
See axolotl config
axolotl version: 0.4.1
adapter: lora
base_model: JackFram/llama-160m
bf16: auto
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
- c3a48bca22943176_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/c3a48bca22943176_train_data.json
type:
field_instruction: keywords
field_output: text
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: null
eval_max_new_tokens: 128
eval_table_size: null
evals_per_epoch: 5
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 4
gradient_checkpointing: false
group_by_length: false
hub_model_id: tuanna08go/a5398e13-1cfc-5486-27db-03400d5e39af
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0001
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 5
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: 50
micro_batch_size: 2
mlflow_experiment_name: /tmp/c3a48bca22943176_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 1
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
saves_per_epoch: 4
sequence_len: 512
special_tokens:
pad_token: </s>
strict: false
tf32: false
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.05
wandb_entity: null
wandb_mode: online
wandb_name: b998dead-53af-4583-bf44-4616d08e8afd
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: b998dead-53af-4583-bf44-4616d08e8afd
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null
a5398e13-1cfc-5486-27db-03400d5e39af
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: 5.5483
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: 0.0001
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- 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: 50
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 0.0002 | 1 | 6.3293 |
| 6.2716 | 0.0017 | 10 | 6.2705 |
| 5.8982 | 0.0034 | 20 | 5.9787 |
| 5.8464 | 0.0052 | 30 | 5.7175 |
| 5.5491 | 0.0069 | 40 | 5.5754 |
| 5.6158 | 0.0086 | 50 | 5.5483 |
Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1