Hugging Face's logo

NJUDeepEngine
/
MeteoRA-llama2-13b

Safetensors
Model card Files
xet
 Community
ParagonLight commited on
Commit
3685bd4
·
1 Parent(s): 6cb6113

upload loras

Browse files
Files changed (28) hide show
  1. llama2_13b_peft/tracking_shuffled_objects/README.md +69 -0
  2. llama2_13b_peft/tracking_shuffled_objects/adapter_config.json +33 -0
  3. llama2_13b_peft/tracking_shuffled_objects/adapter_model.safetensors +3 -0
  4. llama2_13b_peft/tracking_shuffled_objects/all_results.json +11 -0
  5. llama2_13b_peft/tracking_shuffled_objects/eval_results.json +7 -0
  6. llama2_13b_peft/tracking_shuffled_objects/special_tokens_map.json +24 -0
  7. llama2_13b_peft/tracking_shuffled_objects/tokenizer.model +3 -0
  8. llama2_13b_peft/tracking_shuffled_objects/tokenizer_config.json +45 -0
  9. llama2_13b_peft/tracking_shuffled_objects/train_results.json +7 -0
  10. llama2_13b_peft/tracking_shuffled_objects/trainer_log.jsonl +86 -0
  11. llama2_13b_peft/tracking_shuffled_objects/trainer_state.json +622 -0
  12. llama2_13b_peft/tracking_shuffled_objects/training_args.bin +3 -0
  13. llama2_13b_peft/tracking_shuffled_objects/training_eval_loss.png +0 -0
  14. llama2_13b_peft/tracking_shuffled_objects/training_loss.png +0 -0
  15. llama2_13b_peft/vitaminc_fact_verification/README.md +78 -0
  16. llama2_13b_peft/vitaminc_fact_verification/adapter_config.json +33 -0
  17. llama2_13b_peft/vitaminc_fact_verification/adapter_model.safetensors +3 -0
  18. llama2_13b_peft/vitaminc_fact_verification/all_results.json +11 -0
  19. llama2_13b_peft/vitaminc_fact_verification/eval_results.json +7 -0
  20. llama2_13b_peft/vitaminc_fact_verification/special_tokens_map.json +24 -0
  21. llama2_13b_peft/vitaminc_fact_verification/tokenizer.model +3 -0
  22. llama2_13b_peft/vitaminc_fact_verification/tokenizer_config.json +45 -0
  23. llama2_13b_peft/vitaminc_fact_verification/train_results.json +7 -0
  24. llama2_13b_peft/vitaminc_fact_verification/trainer_log.jsonl +80 -0
  25. llama2_13b_peft/vitaminc_fact_verification/trainer_state.json +589 -0
  26. llama2_13b_peft/vitaminc_fact_verification/training_args.bin +3 -0
  27. llama2_13b_peft/vitaminc_fact_verification/training_eval_loss.png +0 -0
  28. llama2_13b_peft/vitaminc_fact_verification/training_loss.png +0 -0
llama2_13b_peft/tracking_shuffled_objects/README.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ library_name: peft
4
+ tags:
5
+ - llama-factory
6
+ - lora
7
+ - generated_from_trainer
8
+ base_model: /data1/model/llama2/meta-llama/Llama2-13b
9
+ model-index:
10
+ - name: tracking_shuffled_objects_no_sys
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
+ # tracking_shuffled_objects_no_sys
18
+
19
+ This model is a fine-tuned version of [/data1/model/llama2/meta-llama/Llama2-13b](https://huggingface.co//data1/model/llama2/meta-llama/Llama2-13b) on the tracking_shuffled_objects_no_sys dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.3650
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 1e-05
41
+ - train_batch_size: 16
42
+ - eval_batch_size: 16
43
+ - seed: 42
44
+ - distributed_type: multi-GPU
45
+ - num_devices: 2
46
+ - total_train_batch_size: 32
47
+ - total_eval_batch_size: 32
48
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
+ - lr_scheduler_type: cosine
50
+ - lr_scheduler_warmup_steps: 20
51
+ - num_epochs: 10.0
52
+
53
+ ### Training results
54
+
55
+ | Training Loss | Epoch | Step | Validation Loss |
56
+ |:-------------:|:-----:|:----:|:---------------:|
57
+ | 0.3492 | 2.5 | 200 | 0.3666 |
58
+ | 0.3626 | 5.0 | 400 | 0.3644 |
59
+ | 0.3471 | 7.5 | 600 | 0.3638 |
60
+ | 0.3618 | 10.0 | 800 | 0.3650 |
61
+
62
+
63
+ ### Framework versions
64
+
65
+ - PEFT 0.9.0
66
+ - Transformers 4.38.2
67
+ - Pytorch 2.2.1
68
+ - Datasets 2.18.0
69
+ - Tokenizers 0.15.2
llama2_13b_peft/tracking_shuffled_objects/adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/data1/model/llama2/meta-llama/Llama2-13b",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 16,
13
+ "lora_dropout": 0.0,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 8,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "k_proj",
23
+ "gate_proj",
24
+ "o_proj",
25
+ "down_proj",
26
+ "v_proj",
27
+ "up_proj",
28
+ "q_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
llama2_13b_peft/tracking_shuffled_objects/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e571846c5465fea18bd6b37d18e92ef0832d8401ffc5d3d656d6325b5fcec9a
3
+ size 125248064
llama2_13b_peft/tracking_shuffled_objects/all_results.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_loss": 0.3649631142616272,
4
+ "eval_runtime": 8.1913,
5
+ "eval_samples_per_second": 54.936,
6
+ "eval_steps_per_second": 1.831,
7
+ "train_loss": 0.5546135276556015,
8
+ "train_runtime": 1438.1753,
9
+ "train_samples_per_second": 17.731,
10
+ "train_steps_per_second": 0.556
11
+ }
llama2_13b_peft/tracking_shuffled_objects/eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_loss": 0.3649631142616272,
4
+ "eval_runtime": 8.1913,
5
+ "eval_samples_per_second": 54.936,
6
+ "eval_steps_per_second": 1.831
7
+ }
llama2_13b_peft/tracking_shuffled_objects/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
llama2_13b_peft/tracking_shuffled_objects/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
llama2_13b_peft/tracking_shuffled_objects/tokenizer_config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": true,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '\\n' }}{% endif %}{% endfor %}",
33
+ "clean_up_tokenization_spaces": false,
34
+ "eos_token": "</s>",
35
+ "legacy": true,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": "</s>",
38
+ "padding_side": "right",
39
+ "sp_model_kwargs": {},
40
+ "spaces_between_special_tokens": false,
41
+ "split_special_tokens": false,
42
+ "tokenizer_class": "LlamaTokenizer",
43
+ "unk_token": "<unk>",
44
+ "use_default_system_prompt": false
45
+ }
llama2_13b_peft/tracking_shuffled_objects/train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "train_loss": 0.5546135276556015,
4
+ "train_runtime": 1438.1753,
5
+ "train_samples_per_second": 17.731,
6
+ "train_steps_per_second": 0.556
7
+ }
llama2_13b_peft/tracking_shuffled_objects/trainer_log.jsonl ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 10, "total_steps": 800, "loss": 4.3605, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5e-06, "epoch": 0.12, "percentage": 1.25, "elapsed_time": "0:00:18", "remaining_time": "0:24:23"}
2
+ {"current_steps": 20, "total_steps": 800, "loss": 4.372, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1e-05, "epoch": 0.25, "percentage": 2.5, "elapsed_time": "0:00:35", "remaining_time": "0:23:11"}
3
+ {"current_steps": 30, "total_steps": 800, "loss": 4.0101, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.995944990857848e-06, "epoch": 0.38, "percentage": 3.75, "elapsed_time": "0:00:52", "remaining_time": "0:22:38"}
4
+ {"current_steps": 40, "total_steps": 800, "loss": 2.9176, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.983786540671052e-06, "epoch": 0.5, "percentage": 5.0, "elapsed_time": "0:01:10", "remaining_time": "0:22:18"}
5
+ {"current_steps": 50, "total_steps": 800, "loss": 1.6643, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.96354437049027e-06, "epoch": 0.62, "percentage": 6.25, "elapsed_time": "0:01:27", "remaining_time": "0:21:54"}
6
+ {"current_steps": 60, "total_steps": 800, "loss": 0.7889, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.935251313189564e-06, "epoch": 0.75, "percentage": 7.5, "elapsed_time": "0:01:44", "remaining_time": "0:21:32"}
7
+ {"current_steps": 70, "total_steps": 800, "loss": 0.4779, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.89895326021134e-06, "epoch": 0.88, "percentage": 8.75, "elapsed_time": "0:02:02", "remaining_time": "0:21:17"}
8
+ {"current_steps": 80, "total_steps": 800, "loss": 0.3952, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.854709087130261e-06, "epoch": 1.0, "percentage": 10.0, "elapsed_time": "0:02:19", "remaining_time": "0:20:59"}
9
+ {"current_steps": 90, "total_steps": 800, "loss": 0.3717, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.802590558156863e-06, "epoch": 1.12, "percentage": 11.25, "elapsed_time": "0:02:37", "remaining_time": "0:20:41"}
10
+ {"current_steps": 100, "total_steps": 800, "loss": 0.3616, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.742682209735727e-06, "epoch": 1.25, "percentage": 12.5, "elapsed_time": "0:02:55", "remaining_time": "0:20:25"}
11
+ {"current_steps": 110, "total_steps": 800, "loss": 0.3795, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.675081213427076e-06, "epoch": 1.38, "percentage": 13.75, "elapsed_time": "0:03:12", "remaining_time": "0:20:06"}
12
+ {"current_steps": 120, "total_steps": 800, "loss": 0.3667, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.599897218294122e-06, "epoch": 1.5, "percentage": 15.0, "elapsed_time": "0:03:29", "remaining_time": "0:19:49"}
13
+ {"current_steps": 130, "total_steps": 800, "loss": 0.3716, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.517252173051912e-06, "epoch": 1.62, "percentage": 16.25, "elapsed_time": "0:03:47", "remaining_time": "0:19:32"}
14
+ {"current_steps": 140, "total_steps": 800, "loss": 0.3572, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.427280128266049e-06, "epoch": 1.75, "percentage": 17.5, "elapsed_time": "0:04:05", "remaining_time": "0:19:15"}
15
+ {"current_steps": 150, "total_steps": 800, "loss": 0.3568, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.330127018922195e-06, "epoch": 1.88, "percentage": 18.75, "elapsed_time": "0:04:22", "remaining_time": "0:18:58"}
16
+ {"current_steps": 160, "total_steps": 800, "loss": 0.3597, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.225950427718974e-06, "epoch": 2.0, "percentage": 20.0, "elapsed_time": "0:04:40", "remaining_time": "0:18:41"}
17
+ {"current_steps": 170, "total_steps": 800, "loss": 0.3586, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.114919329468283e-06, "epoch": 2.12, "percentage": 21.25, "elapsed_time": "0:04:57", "remaining_time": "0:18:23"}
18
+ {"current_steps": 180, "total_steps": 800, "loss": 0.3536, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.997213817017508e-06, "epoch": 2.25, "percentage": 22.5, "elapsed_time": "0:05:15", "remaining_time": "0:18:06"}
19
+ {"current_steps": 190, "total_steps": 800, "loss": 0.3501, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.873024809138272e-06, "epoch": 2.38, "percentage": 23.75, "elapsed_time": "0:05:33", "remaining_time": "0:17:49"}
20
+ {"current_steps": 200, "total_steps": 800, "loss": 0.3492, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.742553740855507e-06, "epoch": 2.5, "percentage": 25.0, "elapsed_time": "0:05:50", "remaining_time": "0:17:31"}
21
+ {"current_steps": 200, "total_steps": 800, "loss": null, "eval_loss": 0.3665711283683777, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 2.5, "percentage": 25.0, "elapsed_time": "0:05:50", "remaining_time": "0:17:31"}
22
+ {"current_steps": 210, "total_steps": 800, "loss": 0.3462, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.606012236719073e-06, "epoch": 2.62, "percentage": 26.25, "elapsed_time": "0:06:16", "remaining_time": "0:17:37"}
23
+ {"current_steps": 220, "total_steps": 800, "loss": 0.3805, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.463621767547998e-06, "epoch": 2.75, "percentage": 27.5, "elapsed_time": "0:06:33", "remaining_time": "0:17:17"}
24
+ {"current_steps": 230, "total_steps": 800, "loss": 0.3636, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.315613291203977e-06, "epoch": 2.88, "percentage": 28.75, "elapsed_time": "0:06:51", "remaining_time": "0:16:59"}
25
+ {"current_steps": 240, "total_steps": 800, "loss": 0.3528, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.162226877976886e-06, "epoch": 3.0, "percentage": 30.0, "elapsed_time": "0:07:09", "remaining_time": "0:16:41"}
26
+ {"current_steps": 250, "total_steps": 800, "loss": 0.3543, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.003711321189895e-06, "epoch": 3.12, "percentage": 31.25, "elapsed_time": "0:07:26", "remaining_time": "0:16:22"}
27
+ {"current_steps": 260, "total_steps": 800, "loss": 0.3509, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.84032373365578e-06, "epoch": 3.25, "percentage": 32.5, "elapsed_time": "0:07:44", "remaining_time": "0:16:04"}
28
+ {"current_steps": 270, "total_steps": 800, "loss": 0.3611, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.672329130639007e-06, "epoch": 3.38, "percentage": 33.75, "elapsed_time": "0:08:01", "remaining_time": "0:15:45"}
29
+ {"current_steps": 280, "total_steps": 800, "loss": 0.3501, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.500000000000001e-06, "epoch": 3.5, "percentage": 35.0, "elapsed_time": "0:08:19", "remaining_time": "0:15:27"}
30
+ {"current_steps": 290, "total_steps": 800, "loss": 0.3551, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.323615860218844e-06, "epoch": 3.62, "percentage": 36.25, "elapsed_time": "0:08:36", "remaining_time": "0:15:08"}
31
+ {"current_steps": 300, "total_steps": 800, "loss": 0.3624, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.143462807015271e-06, "epoch": 3.75, "percentage": 37.5, "elapsed_time": "0:08:54", "remaining_time": "0:14:50"}
32
+ {"current_steps": 310, "total_steps": 800, "loss": 0.3481, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.959833049300376e-06, "epoch": 3.88, "percentage": 38.75, "elapsed_time": "0:09:11", "remaining_time": "0:14:31"}
33
+ {"current_steps": 320, "total_steps": 800, "loss": 0.3547, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.773024435212678e-06, "epoch": 4.0, "percentage": 40.0, "elapsed_time": "0:09:29", "remaining_time": "0:14:13"}
34
+ {"current_steps": 330, "total_steps": 800, "loss": 0.3483, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.583339969007364e-06, "epoch": 4.12, "percentage": 41.25, "elapsed_time": "0:09:46", "remaining_time": "0:13:55"}
35
+ {"current_steps": 340, "total_steps": 800, "loss": 0.3463, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.391087319582264e-06, "epoch": 4.25, "percentage": 42.5, "elapsed_time": "0:10:04", "remaining_time": "0:13:37"}
36
+ {"current_steps": 350, "total_steps": 800, "loss": 0.3453, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.1965783214377895e-06, "epoch": 4.38, "percentage": 43.75, "elapsed_time": "0:10:21", "remaining_time": "0:13:19"}
37
+ {"current_steps": 360, "total_steps": 800, "loss": 0.3473, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.000128468880223e-06, "epoch": 4.5, "percentage": 45.0, "elapsed_time": "0:10:39", "remaining_time": "0:13:01"}
38
+ {"current_steps": 370, "total_steps": 800, "loss": 0.3585, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.8020564042888015e-06, "epoch": 4.62, "percentage": 46.25, "elapsed_time": "0:10:56", "remaining_time": "0:12:43"}
39
+ {"current_steps": 380, "total_steps": 800, "loss": 0.3557, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.6026834012766155e-06, "epoch": 4.75, "percentage": 47.5, "elapsed_time": "0:11:13", "remaining_time": "0:12:24"}
40
+ {"current_steps": 390, "total_steps": 800, "loss": 0.3526, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.402332843583631e-06, "epoch": 4.88, "percentage": 48.75, "elapsed_time": "0:11:31", "remaining_time": "0:12:06"}
41
+ {"current_steps": 400, "total_steps": 800, "loss": 0.3626, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.201329700547077e-06, "epoch": 5.0, "percentage": 50.0, "elapsed_time": "0:11:49", "remaining_time": "0:11:49"}
42
+ {"current_steps": 400, "total_steps": 800, "loss": null, "eval_loss": 0.3643595278263092, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 5.0, "percentage": 50.0, "elapsed_time": "0:11:49", "remaining_time": "0:11:49"}
43
+ {"current_steps": 410, "total_steps": 800, "loss": 0.345, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5e-06, "epoch": 5.12, "percentage": 51.25, "elapsed_time": "0:12:15", "remaining_time": "0:11:39"}
44
+ {"current_steps": 420, "total_steps": 800, "loss": 0.3539, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.798670299452926e-06, "epoch": 5.25, "percentage": 52.5, "elapsed_time": "0:12:32", "remaining_time": "0:11:20"}
45
+ {"current_steps": 430, "total_steps": 800, "loss": 0.3505, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.597667156416371e-06, "epoch": 5.38, "percentage": 53.75, "elapsed_time": "0:12:49", "remaining_time": "0:11:02"}
46
+ {"current_steps": 440, "total_steps": 800, "loss": 0.3558, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.397316598723385e-06, "epoch": 5.5, "percentage": 55.0, "elapsed_time": "0:13:07", "remaining_time": "0:10:44"}
47
+ {"current_steps": 450, "total_steps": 800, "loss": 0.3609, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1979435957111984e-06, "epoch": 5.62, "percentage": 56.25, "elapsed_time": "0:13:24", "remaining_time": "0:10:26"}
48
+ {"current_steps": 460, "total_steps": 800, "loss": 0.3488, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.999871531119779e-06, "epoch": 5.75, "percentage": 57.5, "elapsed_time": "0:13:42", "remaining_time": "0:10:07"}
49
+ {"current_steps": 470, "total_steps": 800, "loss": 0.3484, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.803421678562213e-06, "epoch": 5.88, "percentage": 58.75, "elapsed_time": "0:13:59", "remaining_time": "0:09:49"}
50
+ {"current_steps": 480, "total_steps": 800, "loss": 0.3392, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.6089126804177373e-06, "epoch": 6.0, "percentage": 60.0, "elapsed_time": "0:14:17", "remaining_time": "0:09:31"}
51
+ {"current_steps": 490, "total_steps": 800, "loss": 0.355, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.416660030992639e-06, "epoch": 6.12, "percentage": 61.25, "elapsed_time": "0:14:34", "remaining_time": "0:09:13"}
52
+ {"current_steps": 500, "total_steps": 800, "loss": 0.3463, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.226975564787322e-06, "epoch": 6.25, "percentage": 62.5, "elapsed_time": "0:14:52", "remaining_time": "0:08:55"}
53
+ {"current_steps": 510, "total_steps": 800, "loss": 0.3423, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.040166950699626e-06, "epoch": 6.38, "percentage": 63.75, "elapsed_time": "0:15:09", "remaining_time": "0:08:37"}
54
+ {"current_steps": 520, "total_steps": 800, "loss": 0.3461, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8565371929847286e-06, "epoch": 6.5, "percentage": 65.0, "elapsed_time": "0:15:27", "remaining_time": "0:08:19"}
55
+ {"current_steps": 530, "total_steps": 800, "loss": 0.3496, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6763841397811576e-06, "epoch": 6.62, "percentage": 66.25, "elapsed_time": "0:15:44", "remaining_time": "0:08:01"}
56
+ {"current_steps": 540, "total_steps": 800, "loss": 0.3491, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5000000000000015e-06, "epoch": 6.75, "percentage": 67.5, "elapsed_time": "0:16:02", "remaining_time": "0:07:43"}
57
+ {"current_steps": 550, "total_steps": 800, "loss": 0.3576, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3276708693609947e-06, "epoch": 6.88, "percentage": 68.75, "elapsed_time": "0:16:19", "remaining_time": "0:07:25"}
58
+ {"current_steps": 560, "total_steps": 800, "loss": 0.3498, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.159676266344222e-06, "epoch": 7.0, "percentage": 70.0, "elapsed_time": "0:16:37", "remaining_time": "0:07:07"}
59
+ {"current_steps": 570, "total_steps": 800, "loss": 0.3491, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.996288678810105e-06, "epoch": 7.12, "percentage": 71.25, "elapsed_time": "0:16:55", "remaining_time": "0:06:49"}
60
+ {"current_steps": 580, "total_steps": 800, "loss": 0.3481, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8377731220231144e-06, "epoch": 7.25, "percentage": 72.5, "elapsed_time": "0:17:12", "remaining_time": "0:06:31"}
61
+ {"current_steps": 590, "total_steps": 800, "loss": 0.3573, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6843867087960252e-06, "epoch": 7.38, "percentage": 73.75, "elapsed_time": "0:17:29", "remaining_time": "0:06:13"}
62
+ {"current_steps": 600, "total_steps": 800, "loss": 0.3471, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5363782324520033e-06, "epoch": 7.5, "percentage": 75.0, "elapsed_time": "0:17:47", "remaining_time": "0:05:55"}
63
+ {"current_steps": 600, "total_steps": 800, "loss": null, "eval_loss": 0.3637930452823639, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 7.5, "percentage": 75.0, "elapsed_time": "0:17:47", "remaining_time": "0:05:55"}
64
+ {"current_steps": 610, "total_steps": 800, "loss": 0.3339, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3939877632809279e-06, "epoch": 7.62, "percentage": 76.25, "elapsed_time": "0:18:13", "remaining_time": "0:05:40"}
65
+ {"current_steps": 620, "total_steps": 800, "loss": 0.3583, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.257446259144494e-06, "epoch": 7.75, "percentage": 77.5, "elapsed_time": "0:18:30", "remaining_time": "0:05:22"}
66
+ {"current_steps": 630, "total_steps": 800, "loss": 0.3528, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1269751908617277e-06, "epoch": 7.88, "percentage": 78.75, "elapsed_time": "0:18:48", "remaining_time": "0:05:04"}
67
+ {"current_steps": 640, "total_steps": 800, "loss": 0.3415, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0027861829824953e-06, "epoch": 8.0, "percentage": 80.0, "elapsed_time": "0:19:05", "remaining_time": "0:04:46"}
68
+ {"current_steps": 650, "total_steps": 800, "loss": 0.3465, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.850806705317183e-07, "epoch": 8.12, "percentage": 81.25, "elapsed_time": "0:19:23", "remaining_time": "0:04:28"}
69
+ {"current_steps": 660, "total_steps": 800, "loss": 0.35, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.740495722810271e-07, "epoch": 8.25, "percentage": 82.5, "elapsed_time": "0:19:41", "remaining_time": "0:04:10"}
70
+ {"current_steps": 670, "total_steps": 800, "loss": 0.3725, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.698729810778065e-07, "epoch": 8.38, "percentage": 83.75, "elapsed_time": "0:19:58", "remaining_time": "0:03:52"}
71
+ {"current_steps": 680, "total_steps": 800, "loss": 0.3434, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.727198717339511e-07, "epoch": 8.5, "percentage": 85.0, "elapsed_time": "0:20:16", "remaining_time": "0:03:34"}
72
+ {"current_steps": 690, "total_steps": 800, "loss": 0.3431, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.827478269480895e-07, "epoch": 8.62, "percentage": 86.25, "elapsed_time": "0:20:33", "remaining_time": "0:03:16"}
73
+ {"current_steps": 700, "total_steps": 800, "loss": 0.3382, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.001027817058789e-07, "epoch": 8.75, "percentage": 87.5, "elapsed_time": "0:20:51", "remaining_time": "0:02:58"}
74
+ {"current_steps": 710, "total_steps": 800, "loss": 0.3536, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.2491878657292643e-07, "epoch": 8.88, "percentage": 88.75, "elapsed_time": "0:21:08", "remaining_time": "0:02:40"}
75
+ {"current_steps": 720, "total_steps": 800, "loss": 0.3364, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.573177902642726e-07, "epoch": 9.0, "percentage": 90.0, "elapsed_time": "0:21:26", "remaining_time": "0:02:22"}
76
+ {"current_steps": 730, "total_steps": 800, "loss": 0.3477, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9740944184313882e-07, "epoch": 9.12, "percentage": 91.25, "elapsed_time": "0:21:43", "remaining_time": "0:02:05"}
77
+ {"current_steps": 740, "total_steps": 800, "loss": 0.3449, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4529091286973994e-07, "epoch": 9.25, "percentage": 92.5, "elapsed_time": "0:22:01", "remaining_time": "0:01:47"}
78
+ {"current_steps": 750, "total_steps": 800, "loss": 0.3545, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0104673978866164e-07, "epoch": 9.38, "percentage": 93.75, "elapsed_time": "0:22:18", "remaining_time": "0:01:29"}
79
+ {"current_steps": 760, "total_steps": 800, "loss": 0.3375, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.474868681043578e-08, "epoch": 9.5, "percentage": 95.0, "elapsed_time": "0:22:35", "remaining_time": "0:01:11"}
80
+ {"current_steps": 770, "total_steps": 800, "loss": 0.3447, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.645562950973014e-08, "epoch": 9.62, "percentage": 96.25, "elapsed_time": "0:22:53", "remaining_time": "0:00:53"}
81
+ {"current_steps": 780, "total_steps": 800, "loss": 0.3467, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6213459328950355e-08, "epoch": 9.75, "percentage": 97.5, "elapsed_time": "0:23:10", "remaining_time": "0:00:35"}
82
+ {"current_steps": 790, "total_steps": 800, "loss": 0.342, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.055009142152066e-09, "epoch": 9.88, "percentage": 98.75, "elapsed_time": "0:23:28", "remaining_time": "0:00:17"}
83
+ {"current_steps": 800, "total_steps": 800, "loss": 0.3618, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "0:23:45", "remaining_time": "0:00:00"}
84
+ {"current_steps": 800, "total_steps": 800, "loss": null, "eval_loss": 0.3649631142616272, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "0:23:45", "remaining_time": "0:00:00"}
85
+ {"current_steps": 800, "total_steps": 800, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "0:23:45", "remaining_time": "0:00:00"}
86
+ {"current_steps": 15, "total_steps": 15, "loss": null, "eval_loss": 0.3649631142616272, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "0:24:02", "remaining_time": "0:00:00"}
llama2_13b_peft/tracking_shuffled_objects/trainer_state.json ADDED
@@ -0,0 +1,622 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 10.0,
5
+ "eval_steps": 200,
6
+ "global_step": 800,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.12,
13
+ "grad_norm": 2.8619604110717773,
14
+ "learning_rate": 5e-06,
15
+ "loss": 4.3605,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.25,
20
+ "grad_norm": 3.857790946960449,
21
+ "learning_rate": 1e-05,
22
+ "loss": 4.372,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.38,
27
+ "grad_norm": 5.3878560066223145,
28
+ "learning_rate": 9.995944990857848e-06,
29
+ "loss": 4.0101,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.5,
34
+ "grad_norm": 5.896845817565918,
35
+ "learning_rate": 9.983786540671052e-06,
36
+ "loss": 2.9176,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.62,
41
+ "grad_norm": 3.613201379776001,
42
+ "learning_rate": 9.96354437049027e-06,
43
+ "loss": 1.6643,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.75,
48
+ "grad_norm": 1.5779223442077637,
49
+ "learning_rate": 9.935251313189564e-06,
50
+ "loss": 0.7889,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.88,
55
+ "grad_norm": 0.8221031427383423,
56
+ "learning_rate": 9.89895326021134e-06,
57
+ "loss": 0.4779,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 1.0,
62
+ "grad_norm": 0.5207726955413818,
63
+ "learning_rate": 9.854709087130261e-06,
64
+ "loss": 0.3952,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 1.12,
69
+ "grad_norm": 0.5719926953315735,
70
+ "learning_rate": 9.802590558156863e-06,
71
+ "loss": 0.3717,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 1.25,
76
+ "grad_norm": 0.8804627656936646,
77
+ "learning_rate": 9.742682209735727e-06,
78
+ "loss": 0.3616,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 1.38,
83
+ "grad_norm": 0.4614345133304596,
84
+ "learning_rate": 9.675081213427076e-06,
85
+ "loss": 0.3795,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 1.5,
90
+ "grad_norm": 0.5621177554130554,
91
+ "learning_rate": 9.599897218294122e-06,
92
+ "loss": 0.3667,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 1.62,
97
+ "grad_norm": 0.547697901725769,
98
+ "learning_rate": 9.517252173051912e-06,
99
+ "loss": 0.3716,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 1.75,
104
+ "grad_norm": 0.5013922452926636,
105
+ "learning_rate": 9.427280128266049e-06,
106
+ "loss": 0.3572,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 1.88,
111
+ "grad_norm": 0.4131128489971161,
112
+ "learning_rate": 9.330127018922195e-06,
113
+ "loss": 0.3568,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 2.0,
118
+ "grad_norm": 0.7218930125236511,
119
+ "learning_rate": 9.225950427718974e-06,
120
+ "loss": 0.3597,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 2.12,
125
+ "grad_norm": 0.5658674240112305,
126
+ "learning_rate": 9.114919329468283e-06,
127
+ "loss": 0.3586,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 2.25,
132
+ "grad_norm": 0.37311697006225586,
133
+ "learning_rate": 8.997213817017508e-06,
134
+ "loss": 0.3536,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 2.38,
139
+ "grad_norm": 0.5113855004310608,
140
+ "learning_rate": 8.873024809138272e-06,
141
+ "loss": 0.3501,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 2.5,
146
+ "grad_norm": 0.5142890810966492,
147
+ "learning_rate": 8.742553740855507e-06,
148
+ "loss": 0.3492,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 2.5,
153
+ "eval_loss": 0.3665711283683777,
154
+ "eval_runtime": 8.1974,
155
+ "eval_samples_per_second": 54.896,
156
+ "eval_steps_per_second": 1.83,
157
+ "step": 200
158
+ },
159
+ {
160
+ "epoch": 2.62,
161
+ "grad_norm": 0.35885754227638245,
162
+ "learning_rate": 8.606012236719073e-06,
163
+ "loss": 0.3462,
164
+ "step": 210
165
+ },
166
+ {
167
+ "epoch": 2.75,
168
+ "grad_norm": 0.5794305801391602,
169
+ "learning_rate": 8.463621767547998e-06,
170
+ "loss": 0.3805,
171
+ "step": 220
172
+ },
173
+ {
174
+ "epoch": 2.88,
175
+ "grad_norm": 0.5061661005020142,
176
+ "learning_rate": 8.315613291203977e-06,
177
+ "loss": 0.3636,
178
+ "step": 230
179
+ },
180
+ {
181
+ "epoch": 3.0,
182
+ "grad_norm": 0.4418283700942993,
183
+ "learning_rate": 8.162226877976886e-06,
184
+ "loss": 0.3528,
185
+ "step": 240
186
+ },
187
+ {
188
+ "epoch": 3.12,
189
+ "grad_norm": 0.47025126218795776,
190
+ "learning_rate": 8.003711321189895e-06,
191
+ "loss": 0.3543,
192
+ "step": 250
193
+ },
194
+ {
195
+ "epoch": 3.25,
196
+ "grad_norm": 0.39721566438674927,
197
+ "learning_rate": 7.84032373365578e-06,
198
+ "loss": 0.3509,
199
+ "step": 260
200
+ },
201
+ {
202
+ "epoch": 3.38,
203
+ "grad_norm": 0.5492234230041504,
204
+ "learning_rate": 7.672329130639007e-06,
205
+ "loss": 0.3611,
206
+ "step": 270
207
+ },
208
+ {
209
+ "epoch": 3.5,
210
+ "grad_norm": 0.4932497441768646,
211
+ "learning_rate": 7.500000000000001e-06,
212
+ "loss": 0.3501,
213
+ "step": 280
214
+ },
215
+ {
216
+ "epoch": 3.62,
217
+ "grad_norm": 0.5507996082305908,
218
+ "learning_rate": 7.323615860218844e-06,
219
+ "loss": 0.3551,
220
+ "step": 290
221
+ },
222
+ {
223
+ "epoch": 3.75,
224
+ "grad_norm": 0.5754275321960449,
225
+ "learning_rate": 7.143462807015271e-06,
226
+ "loss": 0.3624,
227
+ "step": 300
228
+ },
229
+ {
230
+ "epoch": 3.88,
231
+ "grad_norm": 0.518661618232727,
232
+ "learning_rate": 6.959833049300376e-06,
233
+ "loss": 0.3481,
234
+ "step": 310
235
+ },
236
+ {
237
+ "epoch": 4.0,
238
+ "grad_norm": 0.5419223308563232,
239
+ "learning_rate": 6.773024435212678e-06,
240
+ "loss": 0.3547,
241
+ "step": 320
242
+ },
243
+ {
244
+ "epoch": 4.12,
245
+ "grad_norm": 0.4590226709842682,
246
+ "learning_rate": 6.583339969007364e-06,
247
+ "loss": 0.3483,
248
+ "step": 330
249
+ },
250
+ {
251
+ "epoch": 4.25,
252
+ "grad_norm": 0.6402090787887573,
253
+ "learning_rate": 6.391087319582264e-06,
254
+ "loss": 0.3463,
255
+ "step": 340
256
+ },
257
+ {
258
+ "epoch": 4.38,
259
+ "grad_norm": 0.4070948362350464,
260
+ "learning_rate": 6.1965783214377895e-06,
261
+ "loss": 0.3453,
262
+ "step": 350
263
+ },
264
+ {
265
+ "epoch": 4.5,
266
+ "grad_norm": 0.4342328608036041,
267
+ "learning_rate": 6.000128468880223e-06,
268
+ "loss": 0.3473,
269
+ "step": 360
270
+ },
271
+ {
272
+ "epoch": 4.62,
273
+ "grad_norm": 0.42583146691322327,
274
+ "learning_rate": 5.8020564042888015e-06,
275
+ "loss": 0.3585,
276
+ "step": 370
277
+ },
278
+ {
279
+ "epoch": 4.75,
280
+ "grad_norm": 0.5823317170143127,
281
+ "learning_rate": 5.6026834012766155e-06,
282
+ "loss": 0.3557,
283
+ "step": 380
284
+ },
285
+ {
286
+ "epoch": 4.88,
287
+ "grad_norm": 0.5292758941650391,
288
+ "learning_rate": 5.402332843583631e-06,
289
+ "loss": 0.3526,
290
+ "step": 390
291
+ },
292
+ {
293
+ "epoch": 5.0,
294
+ "grad_norm": 0.44697287678718567,
295
+ "learning_rate": 5.201329700547077e-06,
296
+ "loss": 0.3626,
297
+ "step": 400
298
+ },
299
+ {
300
+ "epoch": 5.0,
301
+ "eval_loss": 0.3643595278263092,
302
+ "eval_runtime": 8.1855,
303
+ "eval_samples_per_second": 54.975,
304
+ "eval_steps_per_second": 1.833,
305
+ "step": 400
306
+ },
307
+ {
308
+ "epoch": 5.12,
309
+ "grad_norm": 0.44350212812423706,
310
+ "learning_rate": 5e-06,
311
+ "loss": 0.345,
312
+ "step": 410
313
+ },
314
+ {
315
+ "epoch": 5.25,
316
+ "grad_norm": 0.5301347970962524,
317
+ "learning_rate": 4.798670299452926e-06,
318
+ "loss": 0.3539,
319
+ "step": 420
320
+ },
321
+ {
322
+ "epoch": 5.38,
323
+ "grad_norm": 0.5110909938812256,
324
+ "learning_rate": 4.597667156416371e-06,
325
+ "loss": 0.3505,
326
+ "step": 430
327
+ },
328
+ {
329
+ "epoch": 5.5,
330
+ "grad_norm": 0.5509496331214905,
331
+ "learning_rate": 4.397316598723385e-06,
332
+ "loss": 0.3558,
333
+ "step": 440
334
+ },
335
+ {
336
+ "epoch": 5.62,
337
+ "grad_norm": 0.6367918252944946,
338
+ "learning_rate": 4.1979435957111984e-06,
339
+ "loss": 0.3609,
340
+ "step": 450
341
+ },
342
+ {
343
+ "epoch": 5.75,
344
+ "grad_norm": 0.5148094296455383,
345
+ "learning_rate": 3.999871531119779e-06,
346
+ "loss": 0.3488,
347
+ "step": 460
348
+ },
349
+ {
350
+ "epoch": 5.88,
351
+ "grad_norm": 0.634175717830658,
352
+ "learning_rate": 3.803421678562213e-06,
353
+ "loss": 0.3484,
354
+ "step": 470
355
+ },
356
+ {
357
+ "epoch": 6.0,
358
+ "grad_norm": 0.4973353147506714,
359
+ "learning_rate": 3.6089126804177373e-06,
360
+ "loss": 0.3392,
361
+ "step": 480
362
+ },
363
+ {
364
+ "epoch": 6.12,
365
+ "grad_norm": 0.6146537661552429,
366
+ "learning_rate": 3.416660030992639e-06,
367
+ "loss": 0.355,
368
+ "step": 490
369
+ },
370
+ {
371
+ "epoch": 6.25,
372
+ "grad_norm": 0.47262611985206604,
373
+ "learning_rate": 3.226975564787322e-06,
374
+ "loss": 0.3463,
375
+ "step": 500
376
+ },
377
+ {
378
+ "epoch": 6.38,
379
+ "grad_norm": 0.5029013752937317,
380
+ "learning_rate": 3.040166950699626e-06,
381
+ "loss": 0.3423,
382
+ "step": 510
383
+ },
384
+ {
385
+ "epoch": 6.5,
386
+ "grad_norm": 0.5331050753593445,
387
+ "learning_rate": 2.8565371929847286e-06,
388
+ "loss": 0.3461,
389
+ "step": 520
390
+ },
391
+ {
392
+ "epoch": 6.62,
393
+ "grad_norm": 0.5478771924972534,
394
+ "learning_rate": 2.6763841397811576e-06,
395
+ "loss": 0.3496,
396
+ "step": 530
397
+ },
398
+ {
399
+ "epoch": 6.75,
400
+ "grad_norm": 0.4656558632850647,
401
+ "learning_rate": 2.5000000000000015e-06,
402
+ "loss": 0.3491,
403
+ "step": 540
404
+ },
405
+ {
406
+ "epoch": 6.88,
407
+ "grad_norm": 0.6048917174339294,
408
+ "learning_rate": 2.3276708693609947e-06,
409
+ "loss": 0.3576,
410
+ "step": 550
411
+ },
412
+ {
413
+ "epoch": 7.0,
414
+ "grad_norm": 0.46934497356414795,
415
+ "learning_rate": 2.159676266344222e-06,
416
+ "loss": 0.3498,
417
+ "step": 560
418
+ },
419
+ {
420
+ "epoch": 7.12,
421
+ "grad_norm": 0.4495680034160614,
422
+ "learning_rate": 1.996288678810105e-06,
423
+ "loss": 0.3491,
424
+ "step": 570
425
+ },
426
+ {
427
+ "epoch": 7.25,
428
+ "grad_norm": 0.4595155715942383,
429
+ "learning_rate": 1.8377731220231144e-06,
430
+ "loss": 0.3481,
431
+ "step": 580
432
+ },
433
+ {
434
+ "epoch": 7.38,
435
+ "grad_norm": 0.5048754811286926,
436
+ "learning_rate": 1.6843867087960252e-06,
437
+ "loss": 0.3573,
438
+ "step": 590
439
+ },
440
+ {
441
+ "epoch": 7.5,
442
+ "grad_norm": 0.5682963728904724,
443
+ "learning_rate": 1.5363782324520033e-06,
444
+ "loss": 0.3471,
445
+ "step": 600
446
+ },
447
+ {
448
+ "epoch": 7.5,
449
+ "eval_loss": 0.3637930452823639,
450
+ "eval_runtime": 8.1841,
451
+ "eval_samples_per_second": 54.985,
452
+ "eval_steps_per_second": 1.833,
453
+ "step": 600
454
+ },
455
+ {
456
+ "epoch": 7.62,
457
+ "grad_norm": 0.6272332072257996,
458
+ "learning_rate": 1.3939877632809279e-06,
459
+ "loss": 0.3339,
460
+ "step": 610
461
+ },
462
+ {
463
+ "epoch": 7.75,
464
+ "grad_norm": 0.5155446529388428,
465
+ "learning_rate": 1.257446259144494e-06,
466
+ "loss": 0.3583,
467
+ "step": 620
468
+ },
469
+ {
470
+ "epoch": 7.88,
471
+ "grad_norm": 0.546676516532898,
472
+ "learning_rate": 1.1269751908617277e-06,
473
+ "loss": 0.3528,
474
+ "step": 630
475
+ },
476
+ {
477
+ "epoch": 8.0,
478
+ "grad_norm": 0.4784735143184662,
479
+ "learning_rate": 1.0027861829824953e-06,
480
+ "loss": 0.3415,
481
+ "step": 640
482
+ },
483
+ {
484
+ "epoch": 8.12,
485
+ "grad_norm": 0.5068591833114624,
486
+ "learning_rate": 8.850806705317183e-07,
487
+ "loss": 0.3465,
488
+ "step": 650
489
+ },
490
+ {
491
+ "epoch": 8.25,
492
+ "grad_norm": 0.5461397767066956,
493
+ "learning_rate": 7.740495722810271e-07,
494
+ "loss": 0.35,
495
+ "step": 660
496
+ },
497
+ {
498
+ "epoch": 8.38,
499
+ "grad_norm": 0.4632417857646942,
500
+ "learning_rate": 6.698729810778065e-07,
501
+ "loss": 0.3725,
502
+ "step": 670
503
+ },
504
+ {
505
+ "epoch": 8.5,
506
+ "grad_norm": 0.44176045060157776,
507
+ "learning_rate": 5.727198717339511e-07,
508
+ "loss": 0.3434,
509
+ "step": 680
510
+ },
511
+ {
512
+ "epoch": 8.62,
513
+ "grad_norm": 0.46392500400543213,
514
+ "learning_rate": 4.827478269480895e-07,
515
+ "loss": 0.3431,
516
+ "step": 690
517
+ },
518
+ {
519
+ "epoch": 8.75,
520
+ "grad_norm": 0.4535755515098572,
521
+ "learning_rate": 4.001027817058789e-07,
522
+ "loss": 0.3382,
523
+ "step": 700
524
+ },
525
+ {
526
+ "epoch": 8.88,
527
+ "grad_norm": 0.4244369566440582,
528
+ "learning_rate": 3.2491878657292643e-07,
529
+ "loss": 0.3536,
530
+ "step": 710
531
+ },
532
+ {
533
+ "epoch": 9.0,
534
+ "grad_norm": 0.553550124168396,
535
+ "learning_rate": 2.573177902642726e-07,
536
+ "loss": 0.3364,
537
+ "step": 720
538
+ },
539
+ {
540
+ "epoch": 9.12,
541
+ "grad_norm": 0.5430624485015869,
542
+ "learning_rate": 1.9740944184313882e-07,
543
+ "loss": 0.3477,
544
+ "step": 730
545
+ },
546
+ {
547
+ "epoch": 9.25,
548
+ "grad_norm": 0.5077548623085022,
549
+ "learning_rate": 1.4529091286973994e-07,
550
+ "loss": 0.3449,
551
+ "step": 740
552
+ },
553
+ {
554
+ "epoch": 9.38,
555
+ "grad_norm": 0.53578120470047,
556
+ "learning_rate": 1.0104673978866164e-07,
557
+ "loss": 0.3545,
558
+ "step": 750
559
+ },
560
+ {
561
+ "epoch": 9.5,
562
+ "grad_norm": 0.4798441231250763,
563
+ "learning_rate": 6.474868681043578e-08,
564
+ "loss": 0.3375,
565
+ "step": 760
566
+ },
567
+ {
568
+ "epoch": 9.62,
569
+ "grad_norm": 0.5828143954277039,
570
+ "learning_rate": 3.645562950973014e-08,
571
+ "loss": 0.3447,
572
+ "step": 770
573
+ },
574
+ {
575
+ "epoch": 9.75,
576
+ "grad_norm": 0.4779021143913269,
577
+ "learning_rate": 1.6213459328950355e-08,
578
+ "loss": 0.3467,
579
+ "step": 780
580
+ },
581
+ {
582
+ "epoch": 9.88,
583
+ "grad_norm": 0.5191458463668823,
584
+ "learning_rate": 4.055009142152066e-09,
585
+ "loss": 0.342,
586
+ "step": 790
587
+ },
588
+ {
589
+ "epoch": 10.0,
590
+ "grad_norm": 0.5676460862159729,
591
+ "learning_rate": 0.0,
592
+ "loss": 0.3618,
593
+ "step": 800
594
+ },
595
+ {
596
+ "epoch": 10.0,
597
+ "eval_loss": 0.3649631142616272,
598
+ "eval_runtime": 8.1948,
599
+ "eval_samples_per_second": 54.913,
600
+ "eval_steps_per_second": 1.83,
601
+ "step": 800
602
+ },
603
+ {
604
+ "epoch": 10.0,
605
+ "step": 800,
606
+ "total_flos": 5.3765115830756966e+17,
607
+ "train_loss": 0.5546135276556015,
608
+ "train_runtime": 1438.1753,
609
+ "train_samples_per_second": 17.731,
610
+ "train_steps_per_second": 0.556
611
+ }
612
+ ],
613
+ "logging_steps": 10,
614
+ "max_steps": 800,
615
+ "num_input_tokens_seen": 0,
616
+ "num_train_epochs": 10,
617
+ "save_steps": 1000,
618
+ "total_flos": 5.3765115830756966e+17,
619
+ "train_batch_size": 16,
620
+ "trial_name": null,
621
+ "trial_params": null
622
+ }
llama2_13b_peft/tracking_shuffled_objects/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4096e9077e2ad41046184b829d1a95f558bc234f6d09fd3080a9de371fda1b04
3
+ size 5112
llama2_13b_peft/tracking_shuffled_objects/training_eval_loss.png ADDED
llama2_13b_peft/tracking_shuffled_objects/training_loss.png ADDED
llama2_13b_peft/vitaminc_fact_verification/README.md ADDED
@@ -0,0 +1,78 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ library_name: peft
4
+ tags:
5
+ - llama-factory
6
+ - lora
7
+ - generated_from_trainer
8
+ base_model: /data1/model/llama2/meta-llama/Llama2-13b
9
+ model-index:
10
+ - name: vitaminc_fact_verification_no_sys
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
+ # vitaminc_fact_verification_no_sys
18
+
19
+ This model is a fine-tuned version of [/data1/model/llama2/meta-llama/Llama2-13b](https://huggingface.co//data1/model/llama2/meta-llama/Llama2-13b) on the vitaminc_fact_verification_no_sys dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.1413
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 0.0001
41
+ - train_batch_size: 8
42
+ - eval_batch_size: 8
43
+ - seed: 42
44
+ - distributed_type: multi-GPU
45
+ - num_devices: 3
46
+ - total_train_batch_size: 24
47
+ - total_eval_batch_size: 24
48
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
+ - lr_scheduler_type: cosine
50
+ - lr_scheduler_warmup_steps: 20
51
+ - num_epochs: 10.0
52
+
53
+ ### Training results
54
+
55
+ | Training Loss | Epoch | Step | Validation Loss |
56
+ |:-------------:|:-----:|:----:|:---------------:|
57
+ | 0.4369 | 0.03 | 50 | 0.3792 |
58
+ | 0.2145 | 0.06 | 100 | 0.2272 |
59
+ | 0.2037 | 0.1 | 150 | 0.1837 |
60
+ | 0.1913 | 0.13 | 200 | 0.2707 |
61
+ | 0.1781 | 0.16 | 250 | 0.1783 |
62
+ | 0.2087 | 0.19 | 300 | 0.2067 |
63
+ | 0.2029 | 0.23 | 350 | 0.1516 |
64
+ | 0.1081 | 0.26 | 400 | 0.1555 |
65
+ | 0.1619 | 0.29 | 450 | 0.1446 |
66
+ | 0.111 | 0.32 | 500 | 0.1413 |
67
+ | 0.1546 | 0.36 | 550 | 0.1476 |
68
+ | 0.14 | 0.39 | 600 | 0.1437 |
69
+ | 0.1082 | 0.42 | 650 | 0.1483 |
70
+
71
+
72
+ ### Framework versions
73
+
74
+ - PEFT 0.9.0
75
+ - Transformers 4.38.2
76
+ - Pytorch 2.2.1
77
+ - Datasets 2.18.0
78
+ - Tokenizers 0.15.2
llama2_13b_peft/vitaminc_fact_verification/adapter_config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "/data1/model/llama2/meta-llama/Llama2-13b",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 16,
13
+ "lora_dropout": 0.0,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 8,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "v_proj",
23
+ "down_proj",
24
+ "q_proj",
25
+ "up_proj",
26
+ "gate_proj",
27
+ "o_proj",
28
+ "k_proj"
29
+ ],
30
+ "task_type": "CAUSAL_LM",
31
+ "use_dora": false,
32
+ "use_rslora": false
33
+ }
llama2_13b_peft/vitaminc_fact_verification/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8501f5bcb7191ccc6be34271e2cc59743375b55859db909aa1224980444dc23
3
+ size 125248064
llama2_13b_peft/vitaminc_fact_verification/all_results.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.42,
3
+ "eval_loss": 0.1412981152534485,
4
+ "eval_runtime": 63.9749,
5
+ "eval_samples_per_second": 102.556,
6
+ "eval_steps_per_second": 4.283,
7
+ "train_loss": 0.3310180792441735,
8
+ "train_runtime": 1340.7166,
9
+ "train_samples_per_second": 277.27,
10
+ "train_steps_per_second": 11.554
11
+ }
llama2_13b_peft/vitaminc_fact_verification/eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.42,
3
+ "eval_loss": 0.1412981152534485,
4
+ "eval_runtime": 63.9749,
5
+ "eval_samples_per_second": 102.556,
6
+ "eval_steps_per_second": 4.283
7
+ }
llama2_13b_peft/vitaminc_fact_verification/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
llama2_13b_peft/vitaminc_fact_verification/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
llama2_13b_peft/vitaminc_fact_verification/tokenizer_config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": true,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '\\n' }}{% endif %}{% endfor %}",
33
+ "clean_up_tokenization_spaces": false,
34
+ "eos_token": "</s>",
35
+ "legacy": true,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": "</s>",
38
+ "padding_side": "right",
39
+ "sp_model_kwargs": {},
40
+ "spaces_between_special_tokens": false,
41
+ "split_special_tokens": false,
42
+ "tokenizer_class": "LlamaTokenizer",
43
+ "unk_token": "<unk>",
44
+ "use_default_system_prompt": false
45
+ }
llama2_13b_peft/vitaminc_fact_verification/train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.42,
3
+ "train_loss": 0.3310180792441735,
4
+ "train_runtime": 1340.7166,
5
+ "train_samples_per_second": 277.27,
6
+ "train_steps_per_second": 11.554
7
+ }
llama2_13b_peft/vitaminc_fact_verification/trainer_log.jsonl ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 10, "total_steps": 15490, "loss": 7.5989, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5e-05, "epoch": 0.01, "percentage": 0.06, "elapsed_time": "0:00:09", "remaining_time": "3:56:43"}
2
+ {"current_steps": 20, "total_steps": 15490, "loss": 1.8506, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0001, "epoch": 0.01, "percentage": 0.13, "elapsed_time": "0:00:16", "remaining_time": "3:35:58"}
3
+ {"current_steps": 30, "total_steps": 15490, "loss": 0.5371, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.99998968999259e-05, "epoch": 0.02, "percentage": 0.19, "elapsed_time": "0:00:24", "remaining_time": "3:33:50"}
4
+ {"current_steps": 40, "total_steps": 15490, "loss": 0.4742, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.99995876001288e-05, "epoch": 0.03, "percentage": 0.26, "elapsed_time": "0:00:32", "remaining_time": "3:28:00"}
5
+ {"current_steps": 50, "total_steps": 15490, "loss": 0.4369, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.999907210188424e-05, "epoch": 0.03, "percentage": 0.32, "elapsed_time": "0:00:40", "remaining_time": "3:26:14"}
6
+ {"current_steps": 50, "total_steps": 15490, "loss": null, "eval_loss": 0.3792226016521454, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.03, "percentage": 0.32, "elapsed_time": "0:00:40", "remaining_time": "3:26:14"}
7
+ {"current_steps": 60, "total_steps": 15490, "loss": 0.3655, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.999835040731813e-05, "epoch": 0.04, "percentage": 0.39, "elapsed_time": "0:01:51", "remaining_time": "7:58:48"}
8
+ {"current_steps": 70, "total_steps": 15490, "loss": 0.2957, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.999742251940676e-05, "epoch": 0.05, "percentage": 0.45, "elapsed_time": "0:01:59", "remaining_time": "7:18:24"}
9
+ {"current_steps": 80, "total_steps": 15490, "loss": 0.3023, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.999628844197672e-05, "epoch": 0.05, "percentage": 0.52, "elapsed_time": "0:02:06", "remaining_time": "6:46:12"}
10
+ {"current_steps": 90, "total_steps": 15490, "loss": 0.2589, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.999494817970498e-05, "epoch": 0.06, "percentage": 0.58, "elapsed_time": "0:02:14", "remaining_time": "6:22:25"}
11
+ {"current_steps": 100, "total_steps": 15490, "loss": 0.2145, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.999340173811875e-05, "epoch": 0.06, "percentage": 0.65, "elapsed_time": "0:02:21", "remaining_time": "6:03:03"}
12
+ {"current_steps": 100, "total_steps": 15490, "loss": null, "eval_loss": 0.22720637917518616, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.06, "percentage": 0.65, "elapsed_time": "0:02:21", "remaining_time": "6:03:03"}
13
+ {"current_steps": 110, "total_steps": 15490, "loss": 0.2288, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.999164912359557e-05, "epoch": 0.07, "percentage": 0.71, "elapsed_time": "0:03:33", "remaining_time": "8:17:20"}
14
+ {"current_steps": 120, "total_steps": 15490, "loss": 0.2672, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.998969034336327e-05, "epoch": 0.08, "percentage": 0.77, "elapsed_time": "0:03:41", "remaining_time": "7:51:56"}
15
+ {"current_steps": 130, "total_steps": 15490, "loss": 0.2312, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.99875254054998e-05, "epoch": 0.08, "percentage": 0.84, "elapsed_time": "0:03:49", "remaining_time": "7:32:41"}
16
+ {"current_steps": 140, "total_steps": 15490, "loss": 0.1959, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.99851543189334e-05, "epoch": 0.09, "percentage": 0.9, "elapsed_time": "0:03:57", "remaining_time": "7:13:49"}
17
+ {"current_steps": 150, "total_steps": 15490, "loss": 0.2037, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.998257709344246e-05, "epoch": 0.1, "percentage": 0.97, "elapsed_time": "0:04:05", "remaining_time": "6:57:47"}
18
+ {"current_steps": 150, "total_steps": 15490, "loss": null, "eval_loss": 0.1836628019809723, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.1, "percentage": 0.97, "elapsed_time": "0:04:05", "remaining_time": "6:57:47"}
19
+ {"current_steps": 160, "total_steps": 15490, "loss": 0.2029, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.997979373965541e-05, "epoch": 0.1, "percentage": 1.03, "elapsed_time": "0:05:16", "remaining_time": "8:25:22"}
20
+ {"current_steps": 170, "total_steps": 15490, "loss": 0.1685, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.997680426905086e-05, "epoch": 0.11, "percentage": 1.1, "elapsed_time": "0:05:23", "remaining_time": "8:06:23"}
21
+ {"current_steps": 180, "total_steps": 15490, "loss": 0.2626, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.997360869395738e-05, "epoch": 0.12, "percentage": 1.16, "elapsed_time": "0:05:31", "remaining_time": "7:50:28"}
22
+ {"current_steps": 190, "total_steps": 15490, "loss": 0.2029, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.997020702755353e-05, "epoch": 0.12, "percentage": 1.23, "elapsed_time": "0:05:40", "remaining_time": "7:37:25"}
23
+ {"current_steps": 200, "total_steps": 15490, "loss": 0.1913, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.996659928386778e-05, "epoch": 0.13, "percentage": 1.29, "elapsed_time": "0:05:49", "remaining_time": "7:24:58"}
24
+ {"current_steps": 200, "total_steps": 15490, "loss": null, "eval_loss": 0.27072426676750183, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.13, "percentage": 1.29, "elapsed_time": "0:05:49", "remaining_time": "7:24:58"}
25
+ {"current_steps": 210, "total_steps": 15490, "loss": 0.225, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.996278547777851e-05, "epoch": 0.14, "percentage": 1.36, "elapsed_time": "0:07:00", "remaining_time": "8:30:31"}
26
+ {"current_steps": 220, "total_steps": 15490, "loss": 0.2244, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.995876562501383e-05, "epoch": 0.14, "percentage": 1.42, "elapsed_time": "0:07:08", "remaining_time": "8:15:39"}
27
+ {"current_steps": 230, "total_steps": 15490, "loss": 0.1903, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.995453974215163e-05, "epoch": 0.15, "percentage": 1.48, "elapsed_time": "0:07:16", "remaining_time": "8:02:35"}
28
+ {"current_steps": 240, "total_steps": 15490, "loss": 0.1813, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.995010784661948e-05, "epoch": 0.15, "percentage": 1.55, "elapsed_time": "0:07:25", "remaining_time": "7:51:39"}
29
+ {"current_steps": 250, "total_steps": 15490, "loss": 0.1781, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.994546995669454e-05, "epoch": 0.16, "percentage": 1.61, "elapsed_time": "0:07:33", "remaining_time": "7:40:25"}
30
+ {"current_steps": 250, "total_steps": 15490, "loss": null, "eval_loss": 0.17831701040267944, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.16, "percentage": 1.61, "elapsed_time": "0:07:33", "remaining_time": "7:40:25"}
31
+ {"current_steps": 260, "total_steps": 15490, "loss": 0.194, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.994062609150345e-05, "epoch": 0.17, "percentage": 1.68, "elapsed_time": "0:08:45", "remaining_time": "8:32:51"}
32
+ {"current_steps": 270, "total_steps": 15490, "loss": 0.2086, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.993557627102234e-05, "epoch": 0.17, "percentage": 1.74, "elapsed_time": "0:08:52", "remaining_time": "8:20:21"}
33
+ {"current_steps": 280, "total_steps": 15490, "loss": 0.1526, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.993032051607669e-05, "epoch": 0.18, "percentage": 1.81, "elapsed_time": "0:09:00", "remaining_time": "8:09:02"}
34
+ {"current_steps": 290, "total_steps": 15490, "loss": 0.1478, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.992485884834124e-05, "epoch": 0.19, "percentage": 1.87, "elapsed_time": "0:09:07", "remaining_time": "7:58:21"}
35
+ {"current_steps": 300, "total_steps": 15490, "loss": 0.2087, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.991919129033994e-05, "epoch": 0.19, "percentage": 1.94, "elapsed_time": "0:09:15", "remaining_time": "7:48:26"}
36
+ {"current_steps": 300, "total_steps": 15490, "loss": null, "eval_loss": 0.2066759169101715, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.19, "percentage": 1.94, "elapsed_time": "0:09:15", "remaining_time": "7:48:26"}
37
+ {"current_steps": 310, "total_steps": 15490, "loss": 0.1824, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.991331786544577e-05, "epoch": 0.2, "percentage": 2.0, "elapsed_time": "0:10:27", "remaining_time": "8:31:43"}
38
+ {"current_steps": 320, "total_steps": 15490, "loss": 0.1129, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.990723859788081e-05, "epoch": 0.21, "percentage": 2.07, "elapsed_time": "0:10:35", "remaining_time": "8:22:03"}
39
+ {"current_steps": 330, "total_steps": 15490, "loss": 0.1886, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.990095351271594e-05, "epoch": 0.21, "percentage": 2.13, "elapsed_time": "0:10:43", "remaining_time": "8:12:22"}
40
+ {"current_steps": 340, "total_steps": 15490, "loss": 0.1969, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.98944626358709e-05, "epoch": 0.22, "percentage": 2.19, "elapsed_time": "0:10:50", "remaining_time": "8:03:13"}
41
+ {"current_steps": 350, "total_steps": 15490, "loss": 0.2029, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.988776599411405e-05, "epoch": 0.23, "percentage": 2.26, "elapsed_time": "0:10:57", "remaining_time": "7:54:14"}
42
+ {"current_steps": 350, "total_steps": 15490, "loss": null, "eval_loss": 0.15160420536994934, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.23, "percentage": 2.26, "elapsed_time": "0:10:57", "remaining_time": "7:54:14"}
43
+ {"current_steps": 360, "total_steps": 15490, "loss": 0.1585, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.98808636150624e-05, "epoch": 0.23, "percentage": 2.32, "elapsed_time": "0:12:09", "remaining_time": "8:31:04"}
44
+ {"current_steps": 370, "total_steps": 15490, "loss": 0.108, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.987375552718133e-05, "epoch": 0.24, "percentage": 2.39, "elapsed_time": "0:12:17", "remaining_time": "8:22:03"}
45
+ {"current_steps": 380, "total_steps": 15490, "loss": 0.135, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.986644175978466e-05, "epoch": 0.25, "percentage": 2.45, "elapsed_time": "0:12:24", "remaining_time": "8:13:16"}
46
+ {"current_steps": 390, "total_steps": 15490, "loss": 0.1606, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.985892234303438e-05, "epoch": 0.25, "percentage": 2.52, "elapsed_time": "0:12:32", "remaining_time": "8:05:16"}
47
+ {"current_steps": 400, "total_steps": 15490, "loss": 0.1081, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.985119730794056e-05, "epoch": 0.26, "percentage": 2.58, "elapsed_time": "0:12:39", "remaining_time": "7:57:21"}
48
+ {"current_steps": 400, "total_steps": 15490, "loss": null, "eval_loss": 0.15552477538585663, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.26, "percentage": 2.58, "elapsed_time": "0:12:39", "remaining_time": "7:57:21"}
49
+ {"current_steps": 410, "total_steps": 15490, "loss": 0.1763, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.984326668636131e-05, "epoch": 0.26, "percentage": 2.65, "elapsed_time": "0:13:51", "remaining_time": "8:29:40"}
50
+ {"current_steps": 420, "total_steps": 15490, "loss": 0.1041, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.98351305110025e-05, "epoch": 0.27, "percentage": 2.71, "elapsed_time": "0:13:58", "remaining_time": "8:21:39"}
51
+ {"current_steps": 430, "total_steps": 15490, "loss": 0.145, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.982678881541776e-05, "epoch": 0.28, "percentage": 2.78, "elapsed_time": "0:14:06", "remaining_time": "8:14:15"}
52
+ {"current_steps": 440, "total_steps": 15490, "loss": 0.1629, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.981824163400827e-05, "epoch": 0.28, "percentage": 2.84, "elapsed_time": "0:14:14", "remaining_time": "8:07:10"}
53
+ {"current_steps": 450, "total_steps": 15490, "loss": 0.1619, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.980948900202262e-05, "epoch": 0.29, "percentage": 2.91, "elapsed_time": "0:14:22", "remaining_time": "8:00:16"}
54
+ {"current_steps": 450, "total_steps": 15490, "loss": null, "eval_loss": 0.14458192884922028, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.29, "percentage": 2.91, "elapsed_time": "0:14:22", "remaining_time": "8:00:16"}
55
+ {"current_steps": 460, "total_steps": 15490, "loss": 0.136, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.98005309555567e-05, "epoch": 0.3, "percentage": 2.97, "elapsed_time": "0:15:33", "remaining_time": "8:28:26"}
56
+ {"current_steps": 470, "total_steps": 15490, "loss": 0.1194, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.979136753155351e-05, "epoch": 0.3, "percentage": 3.03, "elapsed_time": "0:15:41", "remaining_time": "8:21:33"}
57
+ {"current_steps": 480, "total_steps": 15490, "loss": 0.1498, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.978199876780304e-05, "epoch": 0.31, "percentage": 3.1, "elapsed_time": "0:15:48", "remaining_time": "8:14:31"}
58
+ {"current_steps": 490, "total_steps": 15490, "loss": 0.1561, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.977242470294212e-05, "epoch": 0.32, "percentage": 3.16, "elapsed_time": "0:15:57", "remaining_time": "8:08:20"}
59
+ {"current_steps": 500, "total_steps": 15490, "loss": 0.111, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.97626453764542e-05, "epoch": 0.32, "percentage": 3.23, "elapsed_time": "0:16:04", "remaining_time": "8:02:07"}
60
+ {"current_steps": 500, "total_steps": 15490, "loss": null, "eval_loss": 0.1412981152534485, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.32, "percentage": 3.23, "elapsed_time": "0:16:04", "remaining_time": "8:02:07"}
61
+ {"current_steps": 510, "total_steps": 15490, "loss": 0.1407, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.975266082866923e-05, "epoch": 0.33, "percentage": 3.29, "elapsed_time": "0:17:17", "remaining_time": "8:27:50"}
62
+ {"current_steps": 520, "total_steps": 15490, "loss": 0.1087, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.974247110076355e-05, "epoch": 0.34, "percentage": 3.36, "elapsed_time": "0:17:24", "remaining_time": "8:21:20"}
63
+ {"current_steps": 530, "total_steps": 15490, "loss": 0.1808, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.973207623475965e-05, "epoch": 0.34, "percentage": 3.42, "elapsed_time": "0:17:31", "remaining_time": "8:14:48"}
64
+ {"current_steps": 540, "total_steps": 15490, "loss": 0.1508, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.972147627352592e-05, "epoch": 0.35, "percentage": 3.49, "elapsed_time": "0:17:39", "remaining_time": "8:08:53"}
65
+ {"current_steps": 550, "total_steps": 15490, "loss": 0.1546, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.971067126077669e-05, "epoch": 0.36, "percentage": 3.55, "elapsed_time": "0:17:47", "remaining_time": "8:03:12"}
66
+ {"current_steps": 550, "total_steps": 15490, "loss": null, "eval_loss": 0.14756999909877777, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.36, "percentage": 3.55, "elapsed_time": "0:17:47", "remaining_time": "8:03:12"}
67
+ {"current_steps": 560, "total_steps": 15490, "loss": 0.1303, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.969966124107183e-05, "epoch": 0.36, "percentage": 3.62, "elapsed_time": "0:18:58", "remaining_time": "8:26:03"}
68
+ {"current_steps": 570, "total_steps": 15490, "loss": 0.1683, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.968844625981672e-05, "epoch": 0.37, "percentage": 3.68, "elapsed_time": "0:19:06", "remaining_time": "8:20:21"}
69
+ {"current_steps": 580, "total_steps": 15490, "loss": 0.1199, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.967702636326194e-05, "epoch": 0.37, "percentage": 3.74, "elapsed_time": "0:19:14", "remaining_time": "8:14:39"}
70
+ {"current_steps": 590, "total_steps": 15490, "loss": 0.1296, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.966540159850323e-05, "epoch": 0.38, "percentage": 3.81, "elapsed_time": "0:19:22", "remaining_time": "8:09:06"}
71
+ {"current_steps": 600, "total_steps": 15490, "loss": 0.14, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.965357201348111e-05, "epoch": 0.39, "percentage": 3.87, "elapsed_time": "0:19:29", "remaining_time": "8:03:54"}
72
+ {"current_steps": 600, "total_steps": 15490, "loss": null, "eval_loss": 0.14366257190704346, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.39, "percentage": 3.87, "elapsed_time": "0:19:29", "remaining_time": "8:03:54"}
73
+ {"current_steps": 610, "total_steps": 15490, "loss": 0.1899, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.964153765698083e-05, "epoch": 0.39, "percentage": 3.94, "elapsed_time": "0:20:41", "remaining_time": "8:24:40"}
74
+ {"current_steps": 620, "total_steps": 15490, "loss": 0.1342, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.962929857863215e-05, "epoch": 0.4, "percentage": 4.0, "elapsed_time": "0:20:48", "remaining_time": "8:19:15"}
75
+ {"current_steps": 630, "total_steps": 15490, "loss": 0.1417, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.961685482890901e-05, "epoch": 0.41, "percentage": 4.07, "elapsed_time": "0:20:56", "remaining_time": "8:13:56"}
76
+ {"current_steps": 640, "total_steps": 15490, "loss": 0.1421, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.960420645912949e-05, "epoch": 0.41, "percentage": 4.13, "elapsed_time": "0:21:03", "remaining_time": "8:08:43"}
77
+ {"current_steps": 650, "total_steps": 15490, "loss": 0.1082, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.959135352145552e-05, "epoch": 0.42, "percentage": 4.2, "elapsed_time": "0:21:11", "remaining_time": "8:03:50"}
78
+ {"current_steps": 650, "total_steps": 15490, "loss": null, "eval_loss": 0.14832669496536255, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.42, "percentage": 4.2, "elapsed_time": "0:21:11", "remaining_time": "8:03:50"}
79
+ {"current_steps": 650, "total_steps": 15490, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.42, "percentage": 4.2, "elapsed_time": "0:21:11", "remaining_time": "8:03:50"}
80
+ {"current_steps": 274, "total_steps": 274, "loss": null, "eval_loss": 0.1412981152534485, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.42, "percentage": 100.0, "elapsed_time": "0:23:33", "remaining_time": "0:00:00"}
llama2_13b_peft/vitaminc_fact_verification/trainer_state.json ADDED
@@ -0,0 +1,589 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.1412981152534485,
3
+ "best_model_checkpoint": "ckpt/llama2_13b_fuze15_no_sys/vitaminc_fact_verification_no_sys/checkpoint-500",
4
+ "epoch": 0.4196255648805681,
5
+ "eval_steps": 50,
6
+ "global_step": 650,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "grad_norm": 10.859735488891602,
14
+ "learning_rate": 5e-05,
15
+ "loss": 7.5989,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.01,
20
+ "grad_norm": 18.92371368408203,
21
+ "learning_rate": 0.0001,
22
+ "loss": 1.8506,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.02,
27
+ "grad_norm": 4.927052974700928,
28
+ "learning_rate": 9.99998968999259e-05,
29
+ "loss": 0.5371,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.03,
34
+ "grad_norm": 0.9307419657707214,
35
+ "learning_rate": 9.99995876001288e-05,
36
+ "loss": 0.4742,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.03,
41
+ "grad_norm": 1.7804088592529297,
42
+ "learning_rate": 9.999907210188424e-05,
43
+ "loss": 0.4369,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.03,
48
+ "eval_loss": 0.3792226016521454,
49
+ "eval_runtime": 64.0364,
50
+ "eval_samples_per_second": 102.457,
51
+ "eval_steps_per_second": 4.279,
52
+ "step": 50
53
+ },
54
+ {
55
+ "epoch": 0.04,
56
+ "grad_norm": 1.2824290990829468,
57
+ "learning_rate": 9.999835040731813e-05,
58
+ "loss": 0.3655,
59
+ "step": 60
60
+ },
61
+ {
62
+ "epoch": 0.05,
63
+ "grad_norm": 2.9581339359283447,
64
+ "learning_rate": 9.999742251940676e-05,
65
+ "loss": 0.2957,
66
+ "step": 70
67
+ },
68
+ {
69
+ "epoch": 0.05,
70
+ "grad_norm": 1.5681889057159424,
71
+ "learning_rate": 9.999628844197672e-05,
72
+ "loss": 0.3023,
73
+ "step": 80
74
+ },
75
+ {
76
+ "epoch": 0.06,
77
+ "grad_norm": 2.3982322216033936,
78
+ "learning_rate": 9.999494817970498e-05,
79
+ "loss": 0.2589,
80
+ "step": 90
81
+ },
82
+ {
83
+ "epoch": 0.06,
84
+ "grad_norm": 1.3965446949005127,
85
+ "learning_rate": 9.999340173811875e-05,
86
+ "loss": 0.2145,
87
+ "step": 100
88
+ },
89
+ {
90
+ "epoch": 0.06,
91
+ "eval_loss": 0.22720637917518616,
92
+ "eval_runtime": 64.0931,
93
+ "eval_samples_per_second": 102.367,
94
+ "eval_steps_per_second": 4.275,
95
+ "step": 100
96
+ },
97
+ {
98
+ "epoch": 0.07,
99
+ "grad_norm": 2.2436370849609375,
100
+ "learning_rate": 9.999164912359557e-05,
101
+ "loss": 0.2288,
102
+ "step": 110
103
+ },
104
+ {
105
+ "epoch": 0.08,
106
+ "grad_norm": 1.4689403772354126,
107
+ "learning_rate": 9.998969034336327e-05,
108
+ "loss": 0.2672,
109
+ "step": 120
110
+ },
111
+ {
112
+ "epoch": 0.08,
113
+ "grad_norm": 1.8660027980804443,
114
+ "learning_rate": 9.99875254054998e-05,
115
+ "loss": 0.2312,
116
+ "step": 130
117
+ },
118
+ {
119
+ "epoch": 0.09,
120
+ "grad_norm": 2.51678729057312,
121
+ "learning_rate": 9.99851543189334e-05,
122
+ "loss": 0.1959,
123
+ "step": 140
124
+ },
125
+ {
126
+ "epoch": 0.1,
127
+ "grad_norm": 1.6088509559631348,
128
+ "learning_rate": 9.998257709344246e-05,
129
+ "loss": 0.2037,
130
+ "step": 150
131
+ },
132
+ {
133
+ "epoch": 0.1,
134
+ "eval_loss": 0.1836628019809723,
135
+ "eval_runtime": 64.108,
136
+ "eval_samples_per_second": 102.343,
137
+ "eval_steps_per_second": 4.274,
138
+ "step": 150
139
+ },
140
+ {
141
+ "epoch": 0.1,
142
+ "grad_norm": 1.2034131288528442,
143
+ "learning_rate": 9.997979373965541e-05,
144
+ "loss": 0.2029,
145
+ "step": 160
146
+ },
147
+ {
148
+ "epoch": 0.11,
149
+ "grad_norm": 1.6140587329864502,
150
+ "learning_rate": 9.997680426905086e-05,
151
+ "loss": 0.1685,
152
+ "step": 170
153
+ },
154
+ {
155
+ "epoch": 0.12,
156
+ "grad_norm": 0.7357319593429565,
157
+ "learning_rate": 9.997360869395738e-05,
158
+ "loss": 0.2626,
159
+ "step": 180
160
+ },
161
+ {
162
+ "epoch": 0.12,
163
+ "grad_norm": 1.1027621030807495,
164
+ "learning_rate": 9.997020702755353e-05,
165
+ "loss": 0.2029,
166
+ "step": 190
167
+ },
168
+ {
169
+ "epoch": 0.13,
170
+ "grad_norm": 2.1580281257629395,
171
+ "learning_rate": 9.996659928386778e-05,
172
+ "loss": 0.1913,
173
+ "step": 200
174
+ },
175
+ {
176
+ "epoch": 0.13,
177
+ "eval_loss": 0.27072426676750183,
178
+ "eval_runtime": 64.1275,
179
+ "eval_samples_per_second": 102.312,
180
+ "eval_steps_per_second": 4.273,
181
+ "step": 200
182
+ },
183
+ {
184
+ "epoch": 0.14,
185
+ "grad_norm": 0.9698787927627563,
186
+ "learning_rate": 9.996278547777851e-05,
187
+ "loss": 0.225,
188
+ "step": 210
189
+ },
190
+ {
191
+ "epoch": 0.14,
192
+ "grad_norm": 2.197509765625,
193
+ "learning_rate": 9.995876562501383e-05,
194
+ "loss": 0.2244,
195
+ "step": 220
196
+ },
197
+ {
198
+ "epoch": 0.15,
199
+ "grad_norm": 1.7454187870025635,
200
+ "learning_rate": 9.995453974215163e-05,
201
+ "loss": 0.1903,
202
+ "step": 230
203
+ },
204
+ {
205
+ "epoch": 0.15,
206
+ "grad_norm": 0.8432933688163757,
207
+ "learning_rate": 9.995010784661948e-05,
208
+ "loss": 0.1813,
209
+ "step": 240
210
+ },
211
+ {
212
+ "epoch": 0.16,
213
+ "grad_norm": 0.7707247138023376,
214
+ "learning_rate": 9.994546995669454e-05,
215
+ "loss": 0.1781,
216
+ "step": 250
217
+ },
218
+ {
219
+ "epoch": 0.16,
220
+ "eval_loss": 0.17831701040267944,
221
+ "eval_runtime": 64.1171,
222
+ "eval_samples_per_second": 102.328,
223
+ "eval_steps_per_second": 4.273,
224
+ "step": 250
225
+ },
226
+ {
227
+ "epoch": 0.17,
228
+ "grad_norm": 0.5202718377113342,
229
+ "learning_rate": 9.994062609150345e-05,
230
+ "loss": 0.194,
231
+ "step": 260
232
+ },
233
+ {
234
+ "epoch": 0.17,
235
+ "grad_norm": 1.0719162225723267,
236
+ "learning_rate": 9.993557627102234e-05,
237
+ "loss": 0.2086,
238
+ "step": 270
239
+ },
240
+ {
241
+ "epoch": 0.18,
242
+ "grad_norm": 0.8697193264961243,
243
+ "learning_rate": 9.993032051607669e-05,
244
+ "loss": 0.1526,
245
+ "step": 280
246
+ },
247
+ {
248
+ "epoch": 0.19,
249
+ "grad_norm": 1.0092685222625732,
250
+ "learning_rate": 9.992485884834124e-05,
251
+ "loss": 0.1478,
252
+ "step": 290
253
+ },
254
+ {
255
+ "epoch": 0.19,
256
+ "grad_norm": 2.2846994400024414,
257
+ "learning_rate": 9.991919129033994e-05,
258
+ "loss": 0.2087,
259
+ "step": 300
260
+ },
261
+ {
262
+ "epoch": 0.19,
263
+ "eval_loss": 0.2066759169101715,
264
+ "eval_runtime": 64.114,
265
+ "eval_samples_per_second": 102.333,
266
+ "eval_steps_per_second": 4.274,
267
+ "step": 300
268
+ },
269
+ {
270
+ "epoch": 0.2,
271
+ "grad_norm": 0.825279176235199,
272
+ "learning_rate": 9.991331786544577e-05,
273
+ "loss": 0.1824,
274
+ "step": 310
275
+ },
276
+ {
277
+ "epoch": 0.21,
278
+ "grad_norm": 1.030802845954895,
279
+ "learning_rate": 9.990723859788081e-05,
280
+ "loss": 0.1129,
281
+ "step": 320
282
+ },
283
+ {
284
+ "epoch": 0.21,
285
+ "grad_norm": 1.1893362998962402,
286
+ "learning_rate": 9.990095351271594e-05,
287
+ "loss": 0.1886,
288
+ "step": 330
289
+ },
290
+ {
291
+ "epoch": 0.22,
292
+ "grad_norm": 1.1142666339874268,
293
+ "learning_rate": 9.98944626358709e-05,
294
+ "loss": 0.1969,
295
+ "step": 340
296
+ },
297
+ {
298
+ "epoch": 0.23,
299
+ "grad_norm": 1.058111310005188,
300
+ "learning_rate": 9.988776599411405e-05,
301
+ "loss": 0.2029,
302
+ "step": 350
303
+ },
304
+ {
305
+ "epoch": 0.23,
306
+ "eval_loss": 0.15160420536994934,
307
+ "eval_runtime": 64.19,
308
+ "eval_samples_per_second": 102.212,
309
+ "eval_steps_per_second": 4.269,
310
+ "step": 350
311
+ },
312
+ {
313
+ "epoch": 0.23,
314
+ "grad_norm": 1.1944071054458618,
315
+ "learning_rate": 9.98808636150624e-05,
316
+ "loss": 0.1585,
317
+ "step": 360
318
+ },
319
+ {
320
+ "epoch": 0.24,
321
+ "grad_norm": 0.920417845249176,
322
+ "learning_rate": 9.987375552718133e-05,
323
+ "loss": 0.108,
324
+ "step": 370
325
+ },
326
+ {
327
+ "epoch": 0.25,
328
+ "grad_norm": 0.9066952466964722,
329
+ "learning_rate": 9.986644175978466e-05,
330
+ "loss": 0.135,
331
+ "step": 380
332
+ },
333
+ {
334
+ "epoch": 0.25,
335
+ "grad_norm": 0.9051805138587952,
336
+ "learning_rate": 9.985892234303438e-05,
337
+ "loss": 0.1606,
338
+ "step": 390
339
+ },
340
+ {
341
+ "epoch": 0.26,
342
+ "grad_norm": 0.7989512085914612,
343
+ "learning_rate": 9.985119730794056e-05,
344
+ "loss": 0.1081,
345
+ "step": 400
346
+ },
347
+ {
348
+ "epoch": 0.26,
349
+ "eval_loss": 0.15552477538585663,
350
+ "eval_runtime": 64.1478,
351
+ "eval_samples_per_second": 102.279,
352
+ "eval_steps_per_second": 4.271,
353
+ "step": 400
354
+ },
355
+ {
356
+ "epoch": 0.26,
357
+ "grad_norm": 0.9340474009513855,
358
+ "learning_rate": 9.984326668636131e-05,
359
+ "loss": 0.1763,
360
+ "step": 410
361
+ },
362
+ {
363
+ "epoch": 0.27,
364
+ "grad_norm": 2.8541486263275146,
365
+ "learning_rate": 9.98351305110025e-05,
366
+ "loss": 0.1041,
367
+ "step": 420
368
+ },
369
+ {
370
+ "epoch": 0.28,
371
+ "grad_norm": 0.9724450707435608,
372
+ "learning_rate": 9.982678881541776e-05,
373
+ "loss": 0.145,
374
+ "step": 430
375
+ },
376
+ {
377
+ "epoch": 0.28,
378
+ "grad_norm": 0.9071981906890869,
379
+ "learning_rate": 9.981824163400827e-05,
380
+ "loss": 0.1629,
381
+ "step": 440
382
+ },
383
+ {
384
+ "epoch": 0.29,
385
+ "grad_norm": 1.0815389156341553,
386
+ "learning_rate": 9.980948900202262e-05,
387
+ "loss": 0.1619,
388
+ "step": 450
389
+ },
390
+ {
391
+ "epoch": 0.29,
392
+ "eval_loss": 0.14458192884922028,
393
+ "eval_runtime": 64.1153,
394
+ "eval_samples_per_second": 102.331,
395
+ "eval_steps_per_second": 4.274,
396
+ "step": 450
397
+ },
398
+ {
399
+ "epoch": 0.3,
400
+ "grad_norm": 1.084578275680542,
401
+ "learning_rate": 9.98005309555567e-05,
402
+ "loss": 0.136,
403
+ "step": 460
404
+ },
405
+ {
406
+ "epoch": 0.3,
407
+ "grad_norm": 0.9090166091918945,
408
+ "learning_rate": 9.979136753155351e-05,
409
+ "loss": 0.1194,
410
+ "step": 470
411
+ },
412
+ {
413
+ "epoch": 0.31,
414
+ "grad_norm": 1.7761623859405518,
415
+ "learning_rate": 9.978199876780304e-05,
416
+ "loss": 0.1498,
417
+ "step": 480
418
+ },
419
+ {
420
+ "epoch": 0.32,
421
+ "grad_norm": 0.8033456206321716,
422
+ "learning_rate": 9.977242470294212e-05,
423
+ "loss": 0.1561,
424
+ "step": 490
425
+ },
426
+ {
427
+ "epoch": 0.32,
428
+ "grad_norm": 0.4055212438106537,
429
+ "learning_rate": 9.97626453764542e-05,
430
+ "loss": 0.111,
431
+ "step": 500
432
+ },
433
+ {
434
+ "epoch": 0.32,
435
+ "eval_loss": 0.1412981152534485,
436
+ "eval_runtime": 64.1229,
437
+ "eval_samples_per_second": 102.319,
438
+ "eval_steps_per_second": 4.273,
439
+ "step": 500
440
+ },
441
+ {
442
+ "epoch": 0.33,
443
+ "grad_norm": 2.6774351596832275,
444
+ "learning_rate": 9.975266082866923e-05,
445
+ "loss": 0.1407,
446
+ "step": 510
447
+ },
448
+ {
449
+ "epoch": 0.34,
450
+ "grad_norm": 0.5348044037818909,
451
+ "learning_rate": 9.974247110076355e-05,
452
+ "loss": 0.1087,
453
+ "step": 520
454
+ },
455
+ {
456
+ "epoch": 0.34,
457
+ "grad_norm": 0.6100921034812927,
458
+ "learning_rate": 9.973207623475965e-05,
459
+ "loss": 0.1808,
460
+ "step": 530
461
+ },
462
+ {
463
+ "epoch": 0.35,
464
+ "grad_norm": 0.5292113423347473,
465
+ "learning_rate": 9.972147627352592e-05,
466
+ "loss": 0.1508,
467
+ "step": 540
468
+ },
469
+ {
470
+ "epoch": 0.36,
471
+ "grad_norm": 1.4681833982467651,
472
+ "learning_rate": 9.971067126077669e-05,
473
+ "loss": 0.1546,
474
+ "step": 550
475
+ },
476
+ {
477
+ "epoch": 0.36,
478
+ "eval_loss": 0.14756999909877777,
479
+ "eval_runtime": 64.1644,
480
+ "eval_samples_per_second": 102.253,
481
+ "eval_steps_per_second": 4.27,
482
+ "step": 550
483
+ },
484
+ {
485
+ "epoch": 0.36,
486
+ "grad_norm": 0.851148247718811,
487
+ "learning_rate": 9.969966124107183e-05,
488
+ "loss": 0.1303,
489
+ "step": 560
490
+ },
491
+ {
492
+ "epoch": 0.37,
493
+ "grad_norm": 0.9203129410743713,
494
+ "learning_rate": 9.968844625981672e-05,
495
+ "loss": 0.1683,
496
+ "step": 570
497
+ },
498
+ {
499
+ "epoch": 0.37,
500
+ "grad_norm": 1.1840310096740723,
501
+ "learning_rate": 9.967702636326194e-05,
502
+ "loss": 0.1199,
503
+ "step": 580
504
+ },
505
+ {
506
+ "epoch": 0.38,
507
+ "grad_norm": 0.7599567770957947,
508
+ "learning_rate": 9.966540159850323e-05,
509
+ "loss": 0.1296,
510
+ "step": 590
511
+ },
512
+ {
513
+ "epoch": 0.39,
514
+ "grad_norm": 1.0713849067687988,
515
+ "learning_rate": 9.965357201348111e-05,
516
+ "loss": 0.14,
517
+ "step": 600
518
+ },
519
+ {
520
+ "epoch": 0.39,
521
+ "eval_loss": 0.14366257190704346,
522
+ "eval_runtime": 64.1436,
523
+ "eval_samples_per_second": 102.286,
524
+ "eval_steps_per_second": 4.272,
525
+ "step": 600
526
+ },
527
+ {
528
+ "epoch": 0.39,
529
+ "grad_norm": 0.8163577914237976,
530
+ "learning_rate": 9.964153765698083e-05,
531
+ "loss": 0.1899,
532
+ "step": 610
533
+ },
534
+ {
535
+ "epoch": 0.4,
536
+ "grad_norm": 0.6260606050491333,
537
+ "learning_rate": 9.962929857863215e-05,
538
+ "loss": 0.1342,
539
+ "step": 620
540
+ },
541
+ {
542
+ "epoch": 0.41,
543
+ "grad_norm": 1.3450220823287964,
544
+ "learning_rate": 9.961685482890901e-05,
545
+ "loss": 0.1417,
546
+ "step": 630
547
+ },
548
+ {
549
+ "epoch": 0.41,
550
+ "grad_norm": 2.8518264293670654,
551
+ "learning_rate": 9.960420645912949e-05,
552
+ "loss": 0.1421,
553
+ "step": 640
554
+ },
555
+ {
556
+ "epoch": 0.42,
557
+ "grad_norm": 0.3831160366535187,
558
+ "learning_rate": 9.959135352145552e-05,
559
+ "loss": 0.1082,
560
+ "step": 650
561
+ },
562
+ {
563
+ "epoch": 0.42,
564
+ "eval_loss": 0.14832669496536255,
565
+ "eval_runtime": 64.1294,
566
+ "eval_samples_per_second": 102.309,
567
+ "eval_steps_per_second": 4.273,
568
+ "step": 650
569
+ },
570
+ {
571
+ "epoch": 0.42,
572
+ "step": 650,
573
+ "total_flos": 2.307321824972636e+17,
574
+ "train_loss": 0.3310180792441735,
575
+ "train_runtime": 1340.7166,
576
+ "train_samples_per_second": 277.27,
577
+ "train_steps_per_second": 11.554
578
+ }
579
+ ],
580
+ "logging_steps": 10,
581
+ "max_steps": 15490,
582
+ "num_input_tokens_seen": 0,
583
+ "num_train_epochs": 10,
584
+ "save_steps": 500,
585
+ "total_flos": 2.307321824972636e+17,
586
+ "train_batch_size": 8,
587
+ "trial_name": null,
588
+ "trial_params": null
589
+ }
llama2_13b_peft/vitaminc_fact_verification/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8d870d151be3e535cf16486ceecb60cffba0b4d073161e08569bec709eb0934
3
+ size 5112
llama2_13b_peft/vitaminc_fact_verification/training_eval_loss.png ADDED
llama2_13b_peft/vitaminc_fact_verification/training_loss.png ADDED