Hugging Face's logo

pakawadeep
/
openthaigpt-1.0.0-beta-13b-lora-ctfl-th-gec-100st-augmented

Transformers
TensorBoard
Safetensors
Inference Endpoints
Model card Files Metrics Community
pakawadeep commited on
Commit
39d8501
•
1 Parent(s): 7322a15

Training in progress, step 25

Browse files
Files changed (6) hide show
  1. adapter_config.json +34 -0
  2. adapter_model.safetensors +3 -0
  3. runs/May16_16-26-48_cd0c03bbce77/events.out.tfevents.1715877303.cd0c03bbce77.586.0 +3 -0
  4. runs/May16_17-28-52_cd0c03bbce77/events.out.tfevents.1715881406.cd0c03bbce77.19735.0 +3 -0
  5. training_args.bin +3 -0
  6. training_config.json +93 -0
adapter_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "openthaigpt/openthaigpt-1.0.0-beta-13b-chat-hf",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layer_replication": null,
10
+ "layers_pattern": null,
11
+ "layers_to_transform": null,
12
+ "loftq_config": {},
13
+ "lora_alpha": 32,
14
+ "lora_dropout": 0.1,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 8,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "down_proj",
24
+ "k_proj",
25
+ "gate_proj",
26
+ "v_proj",
27
+ "o_proj",
28
+ "q_proj",
29
+ "up_proj"
30
+ ],
31
+ "task_type": "CAUSAL_LM",
32
+ "use_dora": false,
33
+ "use_rslora": false
34
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34417df016c8dda87ee33942045fc2509dedf52d901937fd4325dee8d29af573
3
+ size 62661424
runs/May16_16-26-48_cd0c03bbce77/events.out.tfevents.1715877303.cd0c03bbce77.586.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2df4f830d02aa80d08c039f65bc0df0ff52fa566987f7bd1010cf779b069fc05
3
+ size 112075
runs/May16_17-28-52_cd0c03bbce77/events.out.tfevents.1715881406.cd0c03bbce77.19735.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97145a9651bc10816cf06bffd013b7b92ee26334105677c4cc25caad76d2d1a0
3
+ size 11668
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8e7fd99e2d3884f2aaee9159ed45ed3b6a2aff5f5d4f1c79788d266e61eed6e
3
+ size 6840
training_config.json ADDED
@@ -0,0 +1,93 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "experiment_key": "base",
3
+ "save_safetensors": true,
4
+ "max_shard_size": "10GB",
5
+ "local_rank": 0,
6
+ "use_gradient_checkpointing": true,
7
+ "trainer_key": "lm",
8
+ "force_fp32": false,
9
+ "force_fp16": false,
10
+ "from_gptq": false,
11
+ "huggingface_hub_token": null,
12
+ "single_gpu": true,
13
+ "master_port": 9994,
14
+ "deepspeed_stage": "stage_2",
15
+ "deepspeed_config_path": null,
16
+ "fsdp_strategy": "",
17
+ "fsdp_offload": true,
18
+ "seed": 42,
19
+ "stabilize": false,
20
+ "norm_fp32": false,
21
+ "path_to_env_file": "./.env",
22
+ "prepare_dataset": true,
23
+ "lora_hub_model_id": null,
24
+ "lora_model_local_path": null,
25
+ "fused_model_local_path": null,
26
+ "fuse_after_training": false,
27
+ "quantization_dataset_id": null,
28
+ "quantization_max_samples": 1024,
29
+ "quantized_model_path": "./quantized_model/",
30
+ "quantized_hub_model_id": null,
31
+ "quantized_hub_private_repo": true,
32
+ "dataset_key": "soda",
33
+ "train_local_path_to_data": "./train.jsonl",
34
+ "eval_local_path_to_data": null,
35
+ "shuffle": true,
36
+ "max_eval_samples": 1000,
37
+ "add_eval_to_train_if_no_path": false,
38
+ "tokenizer_name_or_path": null,
39
+ "tokenizer_use_fast": null,
40
+ "tokenizer_padding_side": null,
41
+ "collator_key": "lm",
42
+ "max_length": 2048,
43
+ "model_name_or_path": "openthaigpt/openthaigpt-1.0.0-beta-13b-chat-hf",
44
+ "push_to_hub_bos_add_bos_token": false,
45
+ "use_flash_attention_2": false,
46
+ "trust_remote_code": false,
47
+ "device_map": null,
48
+ "prepare_model_for_kbit_training": true,
49
+ "offload_folder": null,
50
+ "load_in_8bit": false,
51
+ "load_in_4bit": true,
52
+ "llm_int8_threshold": 6.0,
53
+ "llm_int8_has_fp16_weight": true,
54
+ "bnb_4bit_use_double_quant": true,
55
+ "bnb_4bit_quant_type": "nf4",
56
+ "bnb_quantize_after_model_init": false,
57
+ "gptq_bits": 4,
58
+ "gptq_group_size": 128,
59
+ "gptq_disable_exllama": true,
60
+ "apply_lora": true,
61
+ "lora_rank": 8,
62
+ "lora_alpha": 32,
63
+ "lora_dropout": 0.1,
64
+ "raw_lora_target_modules": "all",
65
+ "output_dir": "./outputs/",
66
+ "per_device_train_batch_size": 2,
67
+ "do_eval": false,
68
+ "per_device_eval_batch_size": null,
69
+ "gradient_accumulation_steps": 2,
70
+ "eval_accumulation_steps": null,
71
+ "eval_delay": 0,
72
+ "eval_steps": 1000,
73
+ "warmup_steps": 10,
74
+ "max_steps": 101,
75
+ "num_train_epochs": 1,
76
+ "learning_rate": 0.0002,
77
+ "max_grad_norm": 1.0,
78
+ "weight_decay": 0.001,
79
+ "label_smoothing_factor": 0.0,
80
+ "logging_steps": 1,
81
+ "save_steps": 25,
82
+ "save_total_limit": 1,
83
+ "optim": "paged_adamw_8bit",
84
+ "push_to_hub": true,
85
+ "hub_model_id": "pakawadeep/openthaigpt-1.0.0-beta-13b-lora-ctfl-th-gec-100st-augmented",
86
+ "hub_private_repo": false,
87
+ "neftune_noise_alpha": null,
88
+ "project_name": null,
89
+ "report_to_wandb": false,
90
+ "wandb_api_key": null,
91
+ "wandb_project": null,
92
+ "wandb_entity": null
93
+ }