You can deploy this model and start earning money today!

Alphatao/
d28a62c4-0691-42e4-a0bc-f275efe25315

Name: d28a62c4-0691-42e4-a0bc-f275efe25315
Availability: InStock

generated_from_trainer

custom_code

NousResearch/CodeLlama-7b-hf-flash

adapter

Views

No views yet

See axolotl config

axolotl version: 0.4.1

yaml

adapter: lora
base_model: NousResearch/CodeLlama-7b-hf-flash
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
  - d88f33407719ca91_train_data.json
  ds_type: json
  format: custom
  path: /workspace/input_data/d88f33407719ca91_train_data.json
  type:
    field_instruction: raw_text
    field_output: clean_text
    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: 400
eval_table_size: null
flash_attention: false
gradient_accumulation_steps: 4
gradient_checkpointing: true
group_by_length: false
hub_model_id: Alphatao/d28a62c4-0691-42e4-a0bc-f275efe25315
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: 4217
micro_batch_size: 2
mlflow_experiment_name: /tmp/d88f33407719ca91_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: 400
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: 0.04
wandb_entity: null
wandb_mode: online
wandb_name: 76417fb1-6aec-4eb2-b27d-be2e384ccafa
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 76417fb1-6aec-4eb2-b27d-be2e384ccafa
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null

d28a62c4-0691-42e4-a0bc-f275efe25315

This model is a fine-tuned version of NousResearch/CodeLlama-7b-hf-flash on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.1286

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: 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: 4217

Training results

Training Loss	Epoch	Step	Validation Loss
0.3332	0.0002	1	0.1905
0.2105	0.0904	400	0.1346
0.0539	0.1808	800	0.1325
0.7387	0.2712	1200	0.1320
0.014	0.3616	1600	0.1310
1.2198	0.4520	2000	0.1309
0.2257	0.5424	2400	0.1297
1.0974	0.6328	2800	0.1292
1.0435	0.7232	3200	0.1288
0.1119	0.8137	3600	0.1286
0.235	0.9041	4000	0.1286

Framework versions

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