SouthMemphis commited on
Commit
2c2ca11
1 Parent(s): 97999db

Upload 13 files

Browse files
README.md ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ library_name: peft
4
+ tags:
5
+ - llama-factory
6
+ - lora
7
+ - generated_from_trainer
8
+ base_model: IlyaGusev/saiga_mistral_7b_merged
9
+ model-index:
10
+ - name: sft
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # sft
18
+
19
+ This model is a fine-tuned version of [IlyaGusev/saiga_mistral_7b_merged](https://huggingface.co/IlyaGusev/saiga_mistral_7b_merged) on the gazeta dataset.
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 5e-05
39
+ - train_batch_size: 2
40
+ - eval_batch_size: 8
41
+ - seed: 42
42
+ - distributed_type: multi-GPU
43
+ - num_devices: 2
44
+ - gradient_accumulation_steps: 2
45
+ - total_train_batch_size: 8
46
+ - total_eval_batch_size: 16
47
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
+ - lr_scheduler_type: cosine
49
+ - num_epochs: 2.0
50
+ - mixed_precision_training: Native AMP
51
+
52
+ ### Training results
53
+
54
+
55
+
56
+ ### Framework versions
57
+
58
+ - PEFT 0.10.0
59
+ - Transformers 4.39.1
60
+ - Pytorch 2.1.0+cu121
61
+ - Datasets 2.18.0
62
+ - Tokenizers 0.15.2
adapter_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "IlyaGusev/saiga_mistral_7b_merged",
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": 16,
14
+ "lora_dropout": 0.0,
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": "unsloth",
22
+ "target_modules": [
23
+ "v_proj",
24
+ "k_proj",
25
+ "down_proj",
26
+ "q_proj",
27
+ "o_proj",
28
+ "gate_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 sha256:61b2dc156877e52a601ed4cc623e4f1a54da269872b76eac61173de9149bdb1c
3
+ size 83945296
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<|im_end|>": 32000,
3
+ "<|im_start|>": 32001
4
+ }
all_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.0,
3
+ "train_loss": 0.888002516746521,
4
+ "train_runtime": 10329.3048,
5
+ "train_samples_per_second": 0.387,
6
+ "train_steps_per_second": 0.048
7
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<unk>",
4
+ "<s>",
5
+ "</s>"
6
+ ],
7
+ "bos_token": {
8
+ "content": "<s>",
9
+ "lstrip": false,
10
+ "normalized": true,
11
+ "rstrip": false,
12
+ "single_word": false
13
+ },
14
+ "eos_token": {
15
+ "content": "</s>",
16
+ "lstrip": false,
17
+ "normalized": true,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "pad_token": {
22
+ "content": "<unk>",
23
+ "lstrip": false,
24
+ "normalized": true,
25
+ "rstrip": false,
26
+ "single_word": false
27
+ },
28
+ "unk_token": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false
34
+ }
35
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
tokenizer_config.json ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": true,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": true,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": true,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": true,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "32000": {
31
+ "content": "<|im_end|>",
32
+ "lstrip": true,
33
+ "normalized": true,
34
+ "rstrip": true,
35
+ "single_word": false,
36
+ "special": false
37
+ },
38
+ "32001": {
39
+ "content": "<|im_start|>",
40
+ "lstrip": true,
41
+ "normalized": true,
42
+ "rstrip": true,
43
+ "single_word": false,
44
+ "special": false
45
+ }
46
+ },
47
+ "additional_special_tokens": [
48
+ "<unk>",
49
+ "<s>",
50
+ "</s>"
51
+ ],
52
+ "bos_token": "<s>",
53
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message + '\\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'Human: ' + content + '\\nAssistant: ' }}{% elif message['role'] == 'assistant' %}{{ content + '</s>' + '\\n' }}{% endif %}{% endfor %}",
54
+ "clean_up_tokenization_spaces": false,
55
+ "eos_token": "</s>",
56
+ "legacy": true,
57
+ "model_max_length": 32768,
58
+ "pad_token": "<unk>",
59
+ "padding_side": "right",
60
+ "sp_model_kwargs": {},
61
+ "spaces_between_special_tokens": false,
62
+ "split_special_tokens": false,
63
+ "tokenizer_class": "LlamaTokenizer",
64
+ "unk_token": "<unk>",
65
+ "use_default_system_prompt": true
66
+ }
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.0,
3
+ "train_loss": 0.888002516746521,
4
+ "train_runtime": 10329.3048,
5
+ "train_samples_per_second": 0.387,
6
+ "train_steps_per_second": 0.048
7
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 10, "total_steps": 500, "loss": 1.1181, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.995066821070679e-05, "epoch": 0.04, "percentage": 2.0, "elapsed_time": "0:03:27", "remaining_time": "2:49:43"}
2
+ {"current_steps": 20, "total_steps": 500, "loss": 1.048, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.980286753286195e-05, "epoch": 0.08, "percentage": 4.0, "elapsed_time": "0:06:48", "remaining_time": "2:43:31"}
3
+ {"current_steps": 30, "total_steps": 500, "loss": 1.0172, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9557181268217227e-05, "epoch": 0.12, "percentage": 6.0, "elapsed_time": "0:10:17", "remaining_time": "2:41:06"}
4
+ {"current_steps": 40, "total_steps": 500, "loss": 1.0621, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9214579028215776e-05, "epoch": 0.16, "percentage": 8.0, "elapsed_time": "0:13:47", "remaining_time": "2:38:37"}
5
+ {"current_steps": 50, "total_steps": 500, "loss": 0.994, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.877641290737884e-05, "epoch": 0.2, "percentage": 10.0, "elapsed_time": "0:17:12", "remaining_time": "2:34:51"}
6
+ {"current_steps": 60, "total_steps": 500, "loss": 1.0483, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.8244412147206284e-05, "epoch": 0.24, "percentage": 12.0, "elapsed_time": "0:20:40", "remaining_time": "2:31:34"}
7
+ {"current_steps": 70, "total_steps": 500, "loss": 1.0077, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.762067631165049e-05, "epoch": 0.28, "percentage": 14.0, "elapsed_time": "0:24:05", "remaining_time": "2:27:57"}
8
+ {"current_steps": 80, "total_steps": 500, "loss": 0.9627, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.690766700109659e-05, "epoch": 0.32, "percentage": 16.0, "elapsed_time": "0:27:23", "remaining_time": "2:23:47"}
9
+ {"current_steps": 90, "total_steps": 500, "loss": 0.9753, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.610819813755038e-05, "epoch": 0.36, "percentage": 18.0, "elapsed_time": "0:30:50", "remaining_time": "2:20:28"}
10
+ {"current_steps": 100, "total_steps": 500, "loss": 1.0424, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.522542485937369e-05, "epoch": 0.4, "percentage": 20.0, "elapsed_time": "0:34:18", "remaining_time": "2:17:14"}
11
+ {"current_steps": 110, "total_steps": 500, "loss": 0.9185, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.426283106939474e-05, "epoch": 0.44, "percentage": 22.0, "elapsed_time": "0:37:39", "remaining_time": "2:13:30"}
12
+ {"current_steps": 120, "total_steps": 500, "loss": 1.0032, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.3224215685535294e-05, "epoch": 0.48, "percentage": 24.0, "elapsed_time": "0:41:07", "remaining_time": "2:10:12"}
13
+ {"current_steps": 130, "total_steps": 500, "loss": 0.995, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.211367764821722e-05, "epoch": 0.52, "percentage": 26.0, "elapsed_time": "0:44:32", "remaining_time": "2:06:46"}
14
+ {"current_steps": 140, "total_steps": 500, "loss": 1.0167, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.093559974371725e-05, "epoch": 0.56, "percentage": 28.0, "elapsed_time": "0:47:53", "remaining_time": "2:03:08"}
15
+ {"current_steps": 150, "total_steps": 500, "loss": 0.9908, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.969463130731183e-05, "epoch": 0.6, "percentage": 30.0, "elapsed_time": "0:51:20", "remaining_time": "1:59:48"}
16
+ {"current_steps": 160, "total_steps": 500, "loss": 0.9903, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.8395669874474915e-05, "epoch": 0.64, "percentage": 32.0, "elapsed_time": "0:54:51", "remaining_time": "1:56:34"}
17
+ {"current_steps": 170, "total_steps": 500, "loss": 1.0031, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.704384185254288e-05, "epoch": 0.68, "percentage": 34.0, "elapsed_time": "0:58:20", "remaining_time": "1:53:15"}
18
+ {"current_steps": 180, "total_steps": 500, "loss": 0.9469, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.564448228912682e-05, "epoch": 0.72, "percentage": 36.0, "elapsed_time": "1:01:43", "remaining_time": "1:49:43"}
19
+ {"current_steps": 190, "total_steps": 500, "loss": 0.9221, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.4203113817116957e-05, "epoch": 0.76, "percentage": 38.0, "elapsed_time": "1:05:11", "remaining_time": "1:46:22"}
20
+ {"current_steps": 200, "total_steps": 500, "loss": 0.9588, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.287466299138262e-05, "epoch": 0.8, "percentage": 40.0, "elapsed_time": "1:08:38", "remaining_time": "1:42:57"}
21
+ {"current_steps": 210, "total_steps": 500, "loss": 0.9762, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.1369268142084556e-05, "epoch": 0.84, "percentage": 42.0, "elapsed_time": "1:12:01", "remaining_time": "1:39:27"}
22
+ {"current_steps": 220, "total_steps": 500, "loss": 0.9577, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9838736701271514e-05, "epoch": 0.88, "percentage": 44.0, "elapsed_time": "1:15:32", "remaining_time": "1:36:08"}
23
+ {"current_steps": 230, "total_steps": 500, "loss": 1.0061, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8289108977307067e-05, "epoch": 0.92, "percentage": 46.0, "elapsed_time": "1:19:01", "remaining_time": "1:32:46"}
24
+ {"current_steps": 240, "total_steps": 500, "loss": 1.0507, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6726500642860154e-05, "epoch": 0.96, "percentage": 48.0, "elapsed_time": "1:22:24", "remaining_time": "1:29:16"}
25
+ {"current_steps": 250, "total_steps": 500, "loss": 0.951, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5157078599138977e-05, "epoch": 1.0, "percentage": 50.0, "elapsed_time": "1:25:48", "remaining_time": "1:25:48"}
26
+ {"current_steps": 260, "total_steps": 500, "loss": 0.8208, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3587036637949388e-05, "epoch": 1.04, "percentage": 52.0, "elapsed_time": "1:29:13", "remaining_time": "1:22:21"}
27
+ {"current_steps": 270, "total_steps": 500, "loss": 0.7953, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2022570997628256e-05, "epoch": 1.08, "percentage": 54.0, "elapsed_time": "1:32:42", "remaining_time": "1:18:58"}
28
+ {"current_steps": 280, "total_steps": 500, "loss": 0.7427, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0469855909321564e-05, "epoch": 1.12, "percentage": 56.0, "elapsed_time": "1:36:10", "remaining_time": "1:15:33"}
29
+ {"current_steps": 290, "total_steps": 500, "loss": 0.763, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.908752507440689e-05, "epoch": 1.16, "percentage": 58.0, "elapsed_time": "1:39:36", "remaining_time": "1:12:08"}
30
+ {"current_steps": 300, "total_steps": 500, "loss": 0.7914, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7573960460574133e-05, "epoch": 1.2, "percentage": 60.0, "elapsed_time": "1:43:03", "remaining_time": "1:08:42"}
31
+ {"current_steps": 310, "total_steps": 500, "loss": 0.7915, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6089703032168733e-05, "epoch": 1.24, "percentage": 62.0, "elapsed_time": "1:46:31", "remaining_time": "1:05:17"}
32
+ {"current_steps": 320, "total_steps": 500, "loss": 0.785, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4640610475167898e-05, "epoch": 1.28, "percentage": 64.0, "elapsed_time": "1:49:58", "remaining_time": "1:01:51"}
33
+ {"current_steps": 330, "total_steps": 500, "loss": 0.7762, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3232401695866687e-05, "epoch": 1.32, "percentage": 66.0, "elapsed_time": "1:53:22", "remaining_time": "0:58:24"}
34
+ {"current_steps": 340, "total_steps": 500, "loss": 0.8098, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1870634250967605e-05, "epoch": 1.36, "percentage": 68.0, "elapsed_time": "1:56:49", "remaining_time": "0:54:58"}
35
+ {"current_steps": 350, "total_steps": 500, "loss": 0.7372, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0560682414443315e-05, "epoch": 1.4, "percentage": 70.0, "elapsed_time": "2:00:16", "remaining_time": "0:51:32"}
36
+ {"current_steps": 360, "total_steps": 500, "loss": 0.8166, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.307715967732491e-06, "epoch": 1.44, "percentage": 72.0, "elapsed_time": "2:03:45", "remaining_time": "0:48:07"}
37
+ {"current_steps": 370, "total_steps": 500, "loss": 0.7472, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.116679796974388e-06, "epoch": 1.48, "percentage": 74.0, "elapsed_time": "2:07:11", "remaining_time": "0:44:41"}
38
+ {"current_steps": 380, "total_steps": 500, "loss": 0.7355, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.992274377802327e-06, "epoch": 1.52, "percentage": 76.0, "elapsed_time": "2:10:29", "remaining_time": "0:41:12"}
39
+ {"current_steps": 390, "total_steps": 500, "loss": 0.7639, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.9389372247138e-06, "epoch": 1.56, "percentage": 78.0, "elapsed_time": "2:13:50", "remaining_time": "0:37:45"}
40
+ {"current_steps": 400, "total_steps": 500, "loss": 0.8115, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.960825378228082e-06, "epoch": 1.6, "percentage": 80.0, "elapsed_time": "2:17:20", "remaining_time": "0:34:20"}
41
+ {"current_steps": 410, "total_steps": 500, "loss": 0.7761, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.061798998946459e-06, "epoch": 1.64, "percentage": 82.0, "elapsed_time": "2:20:49", "remaining_time": "0:30:54"}
42
+ {"current_steps": 420, "total_steps": 500, "loss": 0.805, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.245406133261858e-06, "epoch": 1.68, "percentage": 84.0, "elapsed_time": "2:24:19", "remaining_time": "0:27:29"}
43
+ {"current_steps": 430, "total_steps": 500, "loss": 0.7729, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.514868710840723e-06, "epoch": 1.72, "percentage": 86.0, "elapsed_time": "2:27:43", "remaining_time": "0:24:02"}
44
+ {"current_steps": 440, "total_steps": 500, "loss": 0.7135, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8730698291385518e-06, "epoch": 1.76, "percentage": 88.0, "elapsed_time": "2:31:00", "remaining_time": "0:20:35"}
45
+ {"current_steps": 450, "total_steps": 500, "loss": 0.827, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3225423751313942e-06, "epoch": 1.8, "percentage": 90.0, "elapsed_time": "2:34:20", "remaining_time": "0:17:08"}
46
+ {"current_steps": 460, "total_steps": 500, "loss": 0.7455, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.65459029168153e-07, "epoch": 1.84, "percentage": 92.0, "elapsed_time": "2:37:44", "remaining_time": "0:13:42"}
47
+ {"current_steps": 470, "total_steps": 500, "loss": 0.8043, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.036236903938285e-07, "epoch": 1.88, "percentage": 94.0, "elapsed_time": "2:41:12", "remaining_time": "0:10:17"}
48
+ {"current_steps": 480, "total_steps": 500, "loss": 0.8145, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.384643575837203e-07, "epoch": 1.92, "percentage": 96.0, "elapsed_time": "2:44:38", "remaining_time": "0:06:51"}
49
+ {"current_steps": 490, "total_steps": 500, "loss": 0.7569, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.102749348465165e-08, "epoch": 1.96, "percentage": 98.0, "elapsed_time": "2:48:05", "remaining_time": "0:03:25"}
50
+ {"current_steps": 500, "total_steps": 500, "loss": 0.7338, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.973894904597207e-09, "epoch": 2.0, "percentage": 100.0, "elapsed_time": "2:51:35", "remaining_time": "0:00:00"}
51
+ {"current_steps": 500, "total_steps": 500, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 2.0, "percentage": 100.0, "elapsed_time": "2:51:35", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,380 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.0,
5
+ "eval_steps": 500,
6
+ "global_step": 500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04,
13
+ "grad_norm": 2.072805643081665,
14
+ "learning_rate": 4.995066821070679e-05,
15
+ "loss": 1.1181,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.08,
20
+ "grad_norm": 2.6626899242401123,
21
+ "learning_rate": 4.980286753286195e-05,
22
+ "loss": 1.048,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.12,
27
+ "grad_norm": 3.8898932933807373,
28
+ "learning_rate": 4.9557181268217227e-05,
29
+ "loss": 1.0172,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.16,
34
+ "grad_norm": 2.715027093887329,
35
+ "learning_rate": 4.9214579028215776e-05,
36
+ "loss": 1.0621,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.2,
41
+ "grad_norm": 2.233584403991699,
42
+ "learning_rate": 4.877641290737884e-05,
43
+ "loss": 0.994,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.24,
48
+ "grad_norm": 2.1543967723846436,
49
+ "learning_rate": 4.8244412147206284e-05,
50
+ "loss": 1.0483,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.28,
55
+ "grad_norm": 2.3235280513763428,
56
+ "learning_rate": 4.762067631165049e-05,
57
+ "loss": 1.0077,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.32,
62
+ "grad_norm": 2.6280996799468994,
63
+ "learning_rate": 4.690766700109659e-05,
64
+ "loss": 0.9627,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.36,
69
+ "grad_norm": 2.4375860691070557,
70
+ "learning_rate": 4.610819813755038e-05,
71
+ "loss": 0.9753,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.4,
76
+ "grad_norm": 2.634675979614258,
77
+ "learning_rate": 4.522542485937369e-05,
78
+ "loss": 1.0424,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.44,
83
+ "grad_norm": 2.759512424468994,
84
+ "learning_rate": 4.426283106939474e-05,
85
+ "loss": 0.9185,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.48,
90
+ "grad_norm": 2.3730366230010986,
91
+ "learning_rate": 4.3224215685535294e-05,
92
+ "loss": 1.0032,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.52,
97
+ "grad_norm": 3.0645976066589355,
98
+ "learning_rate": 4.211367764821722e-05,
99
+ "loss": 0.995,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.56,
104
+ "grad_norm": 2.892836093902588,
105
+ "learning_rate": 4.093559974371725e-05,
106
+ "loss": 1.0167,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.6,
111
+ "grad_norm": 2.1179070472717285,
112
+ "learning_rate": 3.969463130731183e-05,
113
+ "loss": 0.9908,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.64,
118
+ "grad_norm": 2.6164073944091797,
119
+ "learning_rate": 3.8395669874474915e-05,
120
+ "loss": 0.9903,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.68,
125
+ "grad_norm": 2.0717203617095947,
126
+ "learning_rate": 3.704384185254288e-05,
127
+ "loss": 1.0031,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.72,
132
+ "grad_norm": 2.2452661991119385,
133
+ "learning_rate": 3.564448228912682e-05,
134
+ "loss": 0.9469,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.76,
139
+ "grad_norm": 2.4002890586853027,
140
+ "learning_rate": 3.4203113817116957e-05,
141
+ "loss": 0.9221,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.8,
146
+ "grad_norm": 2.4350810050964355,
147
+ "learning_rate": 3.287466299138262e-05,
148
+ "loss": 0.9588,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.84,
153
+ "grad_norm": 2.338216781616211,
154
+ "learning_rate": 3.1369268142084556e-05,
155
+ "loss": 0.9762,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.88,
160
+ "grad_norm": 2.716254472732544,
161
+ "learning_rate": 2.9838736701271514e-05,
162
+ "loss": 0.9577,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.92,
167
+ "grad_norm": 2.5938873291015625,
168
+ "learning_rate": 2.8289108977307067e-05,
169
+ "loss": 1.0061,
170
+ "step": 230
171
+ },
172
+ {
173
+ "epoch": 0.96,
174
+ "grad_norm": 2.575774669647217,
175
+ "learning_rate": 2.6726500642860154e-05,
176
+ "loss": 1.0507,
177
+ "step": 240
178
+ },
179
+ {
180
+ "epoch": 1.0,
181
+ "grad_norm": 2.5186479091644287,
182
+ "learning_rate": 2.5157078599138977e-05,
183
+ "loss": 0.951,
184
+ "step": 250
185
+ },
186
+ {
187
+ "epoch": 1.04,
188
+ "grad_norm": 2.128981590270996,
189
+ "learning_rate": 2.3587036637949388e-05,
190
+ "loss": 0.8208,
191
+ "step": 260
192
+ },
193
+ {
194
+ "epoch": 1.08,
195
+ "grad_norm": 2.1169350147247314,
196
+ "learning_rate": 2.2022570997628256e-05,
197
+ "loss": 0.7953,
198
+ "step": 270
199
+ },
200
+ {
201
+ "epoch": 1.12,
202
+ "grad_norm": 2.3190295696258545,
203
+ "learning_rate": 2.0469855909321564e-05,
204
+ "loss": 0.7427,
205
+ "step": 280
206
+ },
207
+ {
208
+ "epoch": 1.16,
209
+ "grad_norm": 2.6951375007629395,
210
+ "learning_rate": 1.908752507440689e-05,
211
+ "loss": 0.763,
212
+ "step": 290
213
+ },
214
+ {
215
+ "epoch": 1.2,
216
+ "grad_norm": 2.6969361305236816,
217
+ "learning_rate": 1.7573960460574133e-05,
218
+ "loss": 0.7914,
219
+ "step": 300
220
+ },
221
+ {
222
+ "epoch": 1.24,
223
+ "grad_norm": 2.5656094551086426,
224
+ "learning_rate": 1.6089703032168733e-05,
225
+ "loss": 0.7915,
226
+ "step": 310
227
+ },
228
+ {
229
+ "epoch": 1.28,
230
+ "grad_norm": 3.075153112411499,
231
+ "learning_rate": 1.4640610475167898e-05,
232
+ "loss": 0.785,
233
+ "step": 320
234
+ },
235
+ {
236
+ "epoch": 1.32,
237
+ "grad_norm": 3.3679215908050537,
238
+ "learning_rate": 1.3232401695866687e-05,
239
+ "loss": 0.7762,
240
+ "step": 330
241
+ },
242
+ {
243
+ "epoch": 1.36,
244
+ "grad_norm": 3.499119281768799,
245
+ "learning_rate": 1.1870634250967605e-05,
246
+ "loss": 0.8098,
247
+ "step": 340
248
+ },
249
+ {
250
+ "epoch": 1.4,
251
+ "grad_norm": 2.9737751483917236,
252
+ "learning_rate": 1.0560682414443315e-05,
253
+ "loss": 0.7372,
254
+ "step": 350
255
+ },
256
+ {
257
+ "epoch": 1.44,
258
+ "grad_norm": 3.3601698875427246,
259
+ "learning_rate": 9.307715967732491e-06,
260
+ "loss": 0.8166,
261
+ "step": 360
262
+ },
263
+ {
264
+ "epoch": 1.48,
265
+ "grad_norm": 3.3548429012298584,
266
+ "learning_rate": 8.116679796974388e-06,
267
+ "loss": 0.7472,
268
+ "step": 370
269
+ },
270
+ {
271
+ "epoch": 1.52,
272
+ "grad_norm": 2.8706016540527344,
273
+ "learning_rate": 6.992274377802327e-06,
274
+ "loss": 0.7355,
275
+ "step": 380
276
+ },
277
+ {
278
+ "epoch": 1.56,
279
+ "grad_norm": 2.9783272743225098,
280
+ "learning_rate": 5.9389372247138e-06,
281
+ "loss": 0.7639,
282
+ "step": 390
283
+ },
284
+ {
285
+ "epoch": 1.6,
286
+ "grad_norm": 3.8318257331848145,
287
+ "learning_rate": 4.960825378228082e-06,
288
+ "loss": 0.8115,
289
+ "step": 400
290
+ },
291
+ {
292
+ "epoch": 1.64,
293
+ "grad_norm": 3.5943961143493652,
294
+ "learning_rate": 4.061798998946459e-06,
295
+ "loss": 0.7761,
296
+ "step": 410
297
+ },
298
+ {
299
+ "epoch": 1.68,
300
+ "grad_norm": 3.3045473098754883,
301
+ "learning_rate": 3.245406133261858e-06,
302
+ "loss": 0.805,
303
+ "step": 420
304
+ },
305
+ {
306
+ "epoch": 1.72,
307
+ "grad_norm": 2.879966974258423,
308
+ "learning_rate": 2.514868710840723e-06,
309
+ "loss": 0.7729,
310
+ "step": 430
311
+ },
312
+ {
313
+ "epoch": 1.76,
314
+ "grad_norm": 4.045947551727295,
315
+ "learning_rate": 1.8730698291385518e-06,
316
+ "loss": 0.7135,
317
+ "step": 440
318
+ },
319
+ {
320
+ "epoch": 1.8,
321
+ "grad_norm": 3.1562774181365967,
322
+ "learning_rate": 1.3225423751313942e-06,
323
+ "loss": 0.827,
324
+ "step": 450
325
+ },
326
+ {
327
+ "epoch": 1.84,
328
+ "grad_norm": 3.1504297256469727,
329
+ "learning_rate": 8.65459029168153e-07,
330
+ "loss": 0.7455,
331
+ "step": 460
332
+ },
333
+ {
334
+ "epoch": 1.88,
335
+ "grad_norm": 3.496607780456543,
336
+ "learning_rate": 5.036236903938285e-07,
337
+ "loss": 0.8043,
338
+ "step": 470
339
+ },
340
+ {
341
+ "epoch": 1.92,
342
+ "grad_norm": 3.2006499767303467,
343
+ "learning_rate": 2.384643575837203e-07,
344
+ "loss": 0.8145,
345
+ "step": 480
346
+ },
347
+ {
348
+ "epoch": 1.96,
349
+ "grad_norm": 3.356722116470337,
350
+ "learning_rate": 7.102749348465165e-08,
351
+ "loss": 0.7569,
352
+ "step": 490
353
+ },
354
+ {
355
+ "epoch": 2.0,
356
+ "grad_norm": 3.4808757305145264,
357
+ "learning_rate": 1.973894904597207e-09,
358
+ "loss": 0.7338,
359
+ "step": 500
360
+ },
361
+ {
362
+ "epoch": 2.0,
363
+ "step": 500,
364
+ "total_flos": 3.216214367225774e+17,
365
+ "train_loss": 0.888002516746521,
366
+ "train_runtime": 10329.3048,
367
+ "train_samples_per_second": 0.387,
368
+ "train_steps_per_second": 0.048
369
+ }
370
+ ],
371
+ "logging_steps": 10,
372
+ "max_steps": 500,
373
+ "num_input_tokens_seen": 0,
374
+ "num_train_epochs": 2,
375
+ "save_steps": 1000,
376
+ "total_flos": 3.216214367225774e+17,
377
+ "train_batch_size": 2,
378
+ "trial_name": null,
379
+ "trial_params": null
380
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c20a0a1e5b0f0cec09e51053429726e4b07c51c324e6d510ac8248a91df88a27
3
+ size 5112
training_loss.png ADDED