upload graphcodebert robust, best f1 score at 0.54 at robust checkpoint 200
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- graphcodebert-rdrop/checkpoint-1600/config.json +28 -0
- graphcodebert-rdrop/checkpoint-1600/merges.txt +0 -0
- graphcodebert-rdrop/checkpoint-1600/model.safetensors +3 -0
- graphcodebert-rdrop/checkpoint-1600/optimizer.pt +3 -0
- graphcodebert-rdrop/checkpoint-1600/rng_state.pth +3 -0
- graphcodebert-rdrop/checkpoint-1600/scaler.pt +3 -0
- graphcodebert-rdrop/checkpoint-1600/scheduler.pt +3 -0
- graphcodebert-rdrop/checkpoint-1600/special_tokens_map.json +51 -0
- graphcodebert-rdrop/checkpoint-1600/tokenizer.json +0 -0
- graphcodebert-rdrop/checkpoint-1600/tokenizer_config.json +58 -0
- graphcodebert-rdrop/checkpoint-1600/trainer_state.json +1175 -0
- graphcodebert-rdrop/checkpoint-1600/training_args.bin +3 -0
- graphcodebert-rdrop/checkpoint-1600/vocab.json +0 -0
- graphcodebert-rdrop/checkpoint-1800/config.json +28 -0
- graphcodebert-rdrop/checkpoint-1800/merges.txt +0 -0
- graphcodebert-rdrop/checkpoint-1800/model.safetensors +3 -0
- graphcodebert-rdrop/checkpoint-1800/optimizer.pt +3 -0
- graphcodebert-rdrop/checkpoint-1800/rng_state.pth +3 -0
- graphcodebert-rdrop/checkpoint-1800/scaler.pt +3 -0
- graphcodebert-rdrop/checkpoint-1800/scheduler.pt +3 -0
- graphcodebert-rdrop/checkpoint-1800/special_tokens_map.json +51 -0
- graphcodebert-rdrop/checkpoint-1800/tokenizer.json +0 -0
- graphcodebert-rdrop/checkpoint-1800/tokenizer_config.json +58 -0
- graphcodebert-rdrop/checkpoint-1800/trainer_state.json +1315 -0
- graphcodebert-rdrop/checkpoint-1800/training_args.bin +3 -0
- graphcodebert-rdrop/checkpoint-1800/vocab.json +0 -0
- graphcodebert-rdrop/checkpoint-2000/config.json +28 -0
- graphcodebert-rdrop/checkpoint-2000/merges.txt +0 -0
- graphcodebert-rdrop/checkpoint-2000/model.safetensors +3 -0
- graphcodebert-rdrop/checkpoint-2000/optimizer.pt +3 -0
- graphcodebert-rdrop/checkpoint-2000/rng_state.pth +3 -0
- graphcodebert-rdrop/checkpoint-2000/scaler.pt +3 -0
- graphcodebert-rdrop/checkpoint-2000/scheduler.pt +3 -0
- graphcodebert-rdrop/checkpoint-2000/special_tokens_map.json +51 -0
- graphcodebert-rdrop/checkpoint-2000/tokenizer.json +0 -0
- graphcodebert-rdrop/checkpoint-2000/tokenizer_config.json +58 -0
- graphcodebert-rdrop/checkpoint-2000/trainer_state.json +1467 -0
- graphcodebert-rdrop/checkpoint-2000/training_args.bin +3 -0
- graphcodebert-rdrop/checkpoint-2000/vocab.json +0 -0
- graphcodebert-rdrop/checkpoint-2200/config.json +28 -0
- graphcodebert-rdrop/checkpoint-2200/merges.txt +0 -0
- graphcodebert-rdrop/checkpoint-2200/model.safetensors +3 -0
- graphcodebert-rdrop/checkpoint-2200/optimizer.pt +3 -0
- graphcodebert-rdrop/checkpoint-2200/rng_state.pth +3 -0
- graphcodebert-rdrop/checkpoint-2200/scaler.pt +3 -0
- graphcodebert-rdrop/checkpoint-2200/scheduler.pt +3 -0
- graphcodebert-rdrop/checkpoint-2200/special_tokens_map.json +51 -0
- graphcodebert-rdrop/checkpoint-2200/tokenizer.json +0 -0
- graphcodebert-rdrop/checkpoint-2200/tokenizer_config.json +58 -0
- graphcodebert-rdrop/checkpoint-2200/trainer_state.json +1607 -0
graphcodebert-rdrop/checkpoint-1600/config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.2,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"gradient_checkpointing": false,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.2,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 514,
|
| 18 |
+
"model_type": "roberta",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"output_past": true,
|
| 22 |
+
"pad_token_id": 1,
|
| 23 |
+
"position_embedding_type": "absolute",
|
| 24 |
+
"transformers_version": "4.56.0",
|
| 25 |
+
"type_vocab_size": 1,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 50265
|
| 28 |
+
}
|
graphcodebert-rdrop/checkpoint-1600/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
graphcodebert-rdrop/checkpoint-1600/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d063dd827bd14bb9ea9021ac3203ca6eeb75950097af9e82d2df0faf2713a046
|
| 3 |
+
size 498612824
|
graphcodebert-rdrop/checkpoint-1600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70b38702a884630526f66b322a7191f4b2f72ecf9f2120ef953460d1da15793f
|
| 3 |
+
size 4741923
|
graphcodebert-rdrop/checkpoint-1600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2e0886230c73a787b8682d7bdf9408209eda92aca36d376ac933517421dc8cef
|
| 3 |
+
size 14645
|
graphcodebert-rdrop/checkpoint-1600/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e4420338904222790575638e536e5d59f25756f58070a9dfc3e73602e7fbef5
|
| 3 |
+
size 1383
|
graphcodebert-rdrop/checkpoint-1600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ee0c2dedc3c352f4004205886d6222fa3ba0cc037e9e0892a04434f44cd5ac9
|
| 3 |
+
size 1465
|
graphcodebert-rdrop/checkpoint-1600/special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
graphcodebert-rdrop/checkpoint-1600/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
graphcodebert-rdrop/checkpoint-1600/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 512,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 56 |
+
"trim_offsets": true,
|
| 57 |
+
"unk_token": "<unk>"
|
| 58 |
+
}
|
graphcodebert-rdrop/checkpoint-1600/trainer_state.json
ADDED
|
@@ -0,0 +1,1175 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1000,
|
| 3 |
+
"best_metric": 0.7657238614598489,
|
| 4 |
+
"best_model_checkpoint": "./output_checkpoints/graphcodebert-rdrop/checkpoint-1000",
|
| 5 |
+
"epoch": 0.1024,
|
| 6 |
+
"eval_steps": 1000,
|
| 7 |
+
"global_step": 1600,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00064,
|
| 14 |
+
"grad_norm": 1.8688431978225708,
|
| 15 |
+
"learning_rate": 3.840409643695328e-08,
|
| 16 |
+
"loss": 0.7168,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.00128,
|
| 21 |
+
"grad_norm": 1.835353970527649,
|
| 22 |
+
"learning_rate": 8.10753147002347e-08,
|
| 23 |
+
"loss": 0.7179,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00192,
|
| 28 |
+
"grad_norm": 1.2541388273239136,
|
| 29 |
+
"learning_rate": 1.2374653296351612e-07,
|
| 30 |
+
"loss": 0.7177,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.00256,
|
| 35 |
+
"grad_norm": 1.1381633281707764,
|
| 36 |
+
"learning_rate": 1.6641775122679754e-07,
|
| 37 |
+
"loss": 0.7312,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0032,
|
| 42 |
+
"grad_norm": 1.6313824653625488,
|
| 43 |
+
"learning_rate": 2.0908896949007894e-07,
|
| 44 |
+
"loss": 0.7167,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.00384,
|
| 49 |
+
"grad_norm": 1.092464566230774,
|
| 50 |
+
"learning_rate": 2.517601877533604e-07,
|
| 51 |
+
"loss": 0.7201,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.00448,
|
| 56 |
+
"grad_norm": 1.164099097251892,
|
| 57 |
+
"learning_rate": 2.944314060166418e-07,
|
| 58 |
+
"loss": 0.7195,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.00512,
|
| 63 |
+
"grad_norm": 1.9283920526504517,
|
| 64 |
+
"learning_rate": 3.371026242799232e-07,
|
| 65 |
+
"loss": 0.7223,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.00576,
|
| 70 |
+
"grad_norm": 1.8716129064559937,
|
| 71 |
+
"learning_rate": 3.7977384254320464e-07,
|
| 72 |
+
"loss": 0.7151,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0064,
|
| 77 |
+
"grad_norm": 1.5499262809753418,
|
| 78 |
+
"learning_rate": 4.22445060806486e-07,
|
| 79 |
+
"loss": 0.7211,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.00704,
|
| 84 |
+
"grad_norm": 1.375114917755127,
|
| 85 |
+
"learning_rate": 4.651162790697675e-07,
|
| 86 |
+
"loss": 0.7247,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.00768,
|
| 91 |
+
"grad_norm": 1.384915828704834,
|
| 92 |
+
"learning_rate": 5.077874973330489e-07,
|
| 93 |
+
"loss": 0.7156,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.00832,
|
| 98 |
+
"grad_norm": 1.6376659870147705,
|
| 99 |
+
"learning_rate": 5.504587155963304e-07,
|
| 100 |
+
"loss": 0.7196,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.00896,
|
| 105 |
+
"grad_norm": 1.709489345550537,
|
| 106 |
+
"learning_rate": 5.931299338596117e-07,
|
| 107 |
+
"loss": 0.7215,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0096,
|
| 112 |
+
"grad_norm": 1.350512146949768,
|
| 113 |
+
"learning_rate": 6.358011521228932e-07,
|
| 114 |
+
"loss": 0.71,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.01024,
|
| 119 |
+
"grad_norm": 2.33050537109375,
|
| 120 |
+
"learning_rate": 6.784723703861745e-07,
|
| 121 |
+
"loss": 0.7191,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.01088,
|
| 126 |
+
"grad_norm": 1.0042874813079834,
|
| 127 |
+
"learning_rate": 7.21143588649456e-07,
|
| 128 |
+
"loss": 0.72,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.01152,
|
| 133 |
+
"grad_norm": 1.1835744380950928,
|
| 134 |
+
"learning_rate": 7.638148069127374e-07,
|
| 135 |
+
"loss": 0.7122,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.01216,
|
| 140 |
+
"grad_norm": 1.949506402015686,
|
| 141 |
+
"learning_rate": 8.064860251760189e-07,
|
| 142 |
+
"loss": 0.7091,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.0128,
|
| 147 |
+
"grad_norm": 1.139626383781433,
|
| 148 |
+
"learning_rate": 8.491572434393003e-07,
|
| 149 |
+
"loss": 0.7046,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.01344,
|
| 154 |
+
"grad_norm": 1.7734779119491577,
|
| 155 |
+
"learning_rate": 8.918284617025817e-07,
|
| 156 |
+
"loss": 0.7135,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.01408,
|
| 161 |
+
"grad_norm": 1.3442974090576172,
|
| 162 |
+
"learning_rate": 9.344996799658632e-07,
|
| 163 |
+
"loss": 0.7127,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.01472,
|
| 168 |
+
"grad_norm": 1.6148335933685303,
|
| 169 |
+
"learning_rate": 9.771708982291445e-07,
|
| 170 |
+
"loss": 0.7168,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.01536,
|
| 175 |
+
"grad_norm": 0.9220213294029236,
|
| 176 |
+
"learning_rate": 1.0198421164924258e-06,
|
| 177 |
+
"loss": 0.7109,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.016,
|
| 182 |
+
"grad_norm": 1.5631815195083618,
|
| 183 |
+
"learning_rate": 1.0625133347557074e-06,
|
| 184 |
+
"loss": 0.7149,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.01664,
|
| 189 |
+
"grad_norm": 1.5052822828292847,
|
| 190 |
+
"learning_rate": 1.1051845530189888e-06,
|
| 191 |
+
"loss": 0.7112,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.01728,
|
| 196 |
+
"grad_norm": 1.6714903116226196,
|
| 197 |
+
"learning_rate": 1.1478557712822702e-06,
|
| 198 |
+
"loss": 0.7128,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.01792,
|
| 203 |
+
"grad_norm": 1.6667262315750122,
|
| 204 |
+
"learning_rate": 1.1905269895455517e-06,
|
| 205 |
+
"loss": 0.7091,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.01856,
|
| 210 |
+
"grad_norm": 1.6861907243728638,
|
| 211 |
+
"learning_rate": 1.233198207808833e-06,
|
| 212 |
+
"loss": 0.7078,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0192,
|
| 217 |
+
"grad_norm": 1.6259212493896484,
|
| 218 |
+
"learning_rate": 1.2758694260721145e-06,
|
| 219 |
+
"loss": 0.7087,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.01984,
|
| 224 |
+
"grad_norm": 1.207320213317871,
|
| 225 |
+
"learning_rate": 1.318540644335396e-06,
|
| 226 |
+
"loss": 0.7087,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.02048,
|
| 231 |
+
"grad_norm": 1.087124228477478,
|
| 232 |
+
"learning_rate": 1.3612118625986772e-06,
|
| 233 |
+
"loss": 0.7174,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.02112,
|
| 238 |
+
"grad_norm": 1.5668721199035645,
|
| 239 |
+
"learning_rate": 1.4038830808619588e-06,
|
| 240 |
+
"loss": 0.7066,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.02176,
|
| 245 |
+
"grad_norm": 1.5332859754562378,
|
| 246 |
+
"learning_rate": 1.4465542991252401e-06,
|
| 247 |
+
"loss": 0.7115,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0224,
|
| 252 |
+
"grad_norm": 2.092994451522827,
|
| 253 |
+
"learning_rate": 1.4892255173885215e-06,
|
| 254 |
+
"loss": 0.7085,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.02304,
|
| 259 |
+
"grad_norm": 1.266851544380188,
|
| 260 |
+
"learning_rate": 1.531896735651803e-06,
|
| 261 |
+
"loss": 0.7091,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.02368,
|
| 266 |
+
"grad_norm": 1.290616512298584,
|
| 267 |
+
"learning_rate": 1.5745679539150842e-06,
|
| 268 |
+
"loss": 0.7103,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.02432,
|
| 273 |
+
"grad_norm": 1.2045000791549683,
|
| 274 |
+
"learning_rate": 1.6172391721783658e-06,
|
| 275 |
+
"loss": 0.705,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.02496,
|
| 280 |
+
"grad_norm": 1.828832983970642,
|
| 281 |
+
"learning_rate": 1.6599103904416472e-06,
|
| 282 |
+
"loss": 0.7038,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.0256,
|
| 287 |
+
"grad_norm": 1.7793196439743042,
|
| 288 |
+
"learning_rate": 1.7025816087049288e-06,
|
| 289 |
+
"loss": 0.7085,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.02624,
|
| 294 |
+
"grad_norm": 0.8949472904205322,
|
| 295 |
+
"learning_rate": 1.7452528269682101e-06,
|
| 296 |
+
"loss": 0.7045,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.02688,
|
| 301 |
+
"grad_norm": 1.3071945905685425,
|
| 302 |
+
"learning_rate": 1.7879240452314913e-06,
|
| 303 |
+
"loss": 0.702,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.02752,
|
| 308 |
+
"grad_norm": 1.7497148513793945,
|
| 309 |
+
"learning_rate": 1.8305952634947729e-06,
|
| 310 |
+
"loss": 0.7073,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.02816,
|
| 315 |
+
"grad_norm": 1.0856297016143799,
|
| 316 |
+
"learning_rate": 1.8732664817580542e-06,
|
| 317 |
+
"loss": 0.7008,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.0288,
|
| 322 |
+
"grad_norm": 1.369019627571106,
|
| 323 |
+
"learning_rate": 1.915937700021336e-06,
|
| 324 |
+
"loss": 0.7059,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.02944,
|
| 329 |
+
"grad_norm": 2.0961010456085205,
|
| 330 |
+
"learning_rate": 1.958608918284617e-06,
|
| 331 |
+
"loss": 0.6952,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.03008,
|
| 336 |
+
"grad_norm": 1.6076347827911377,
|
| 337 |
+
"learning_rate": 2.0012801365478988e-06,
|
| 338 |
+
"loss": 0.7074,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.03072,
|
| 343 |
+
"grad_norm": 1.3129311800003052,
|
| 344 |
+
"learning_rate": 2.04395135481118e-06,
|
| 345 |
+
"loss": 0.6994,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.03136,
|
| 350 |
+
"grad_norm": 1.3621476888656616,
|
| 351 |
+
"learning_rate": 2.0866225730744615e-06,
|
| 352 |
+
"loss": 0.696,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.032,
|
| 357 |
+
"grad_norm": 0.9786806106567383,
|
| 358 |
+
"learning_rate": 2.129293791337743e-06,
|
| 359 |
+
"loss": 0.7011,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.03264,
|
| 364 |
+
"grad_norm": 1.4611176252365112,
|
| 365 |
+
"learning_rate": 2.1719650096010242e-06,
|
| 366 |
+
"loss": 0.699,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.03328,
|
| 371 |
+
"grad_norm": 1.0675945281982422,
|
| 372 |
+
"learning_rate": 2.214636227864306e-06,
|
| 373 |
+
"loss": 0.6906,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.03392,
|
| 378 |
+
"grad_norm": 1.6564017534255981,
|
| 379 |
+
"learning_rate": 2.257307446127587e-06,
|
| 380 |
+
"loss": 0.6923,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.03456,
|
| 385 |
+
"grad_norm": 1.228119134902954,
|
| 386 |
+
"learning_rate": 2.2999786643908685e-06,
|
| 387 |
+
"loss": 0.6928,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.0352,
|
| 392 |
+
"grad_norm": 1.9868593215942383,
|
| 393 |
+
"learning_rate": 2.34264988265415e-06,
|
| 394 |
+
"loss": 0.6912,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.03584,
|
| 399 |
+
"grad_norm": 1.2531176805496216,
|
| 400 |
+
"learning_rate": 2.3853211009174317e-06,
|
| 401 |
+
"loss": 0.6923,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.03648,
|
| 406 |
+
"grad_norm": 1.413602352142334,
|
| 407 |
+
"learning_rate": 2.427992319180713e-06,
|
| 408 |
+
"loss": 0.6993,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.03712,
|
| 413 |
+
"grad_norm": 0.9655390977859497,
|
| 414 |
+
"learning_rate": 2.470663537443994e-06,
|
| 415 |
+
"loss": 0.7017,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.03776,
|
| 420 |
+
"grad_norm": 1.962438941001892,
|
| 421 |
+
"learning_rate": 2.5133347557072756e-06,
|
| 422 |
+
"loss": 0.6933,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.0384,
|
| 427 |
+
"grad_norm": 1.1099931001663208,
|
| 428 |
+
"learning_rate": 2.556005973970557e-06,
|
| 429 |
+
"loss": 0.6925,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.03904,
|
| 434 |
+
"grad_norm": 1.1766624450683594,
|
| 435 |
+
"learning_rate": 2.5986771922338383e-06,
|
| 436 |
+
"loss": 0.6894,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.03968,
|
| 441 |
+
"grad_norm": 1.574353814125061,
|
| 442 |
+
"learning_rate": 2.64134841049712e-06,
|
| 443 |
+
"loss": 0.6923,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.04032,
|
| 448 |
+
"grad_norm": 1.194074273109436,
|
| 449 |
+
"learning_rate": 2.6840196287604015e-06,
|
| 450 |
+
"loss": 0.6855,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.04096,
|
| 455 |
+
"grad_norm": 1.1750593185424805,
|
| 456 |
+
"learning_rate": 2.7266908470236826e-06,
|
| 457 |
+
"loss": 0.7002,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.0416,
|
| 462 |
+
"grad_norm": 1.3165347576141357,
|
| 463 |
+
"learning_rate": 2.7693620652869642e-06,
|
| 464 |
+
"loss": 0.6893,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.04224,
|
| 469 |
+
"grad_norm": 1.1079384088516235,
|
| 470 |
+
"learning_rate": 2.812033283550246e-06,
|
| 471 |
+
"loss": 0.6852,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.04288,
|
| 476 |
+
"grad_norm": 1.231327772140503,
|
| 477 |
+
"learning_rate": 2.8547045018135274e-06,
|
| 478 |
+
"loss": 0.695,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.04352,
|
| 483 |
+
"grad_norm": 1.966036319732666,
|
| 484 |
+
"learning_rate": 2.897375720076808e-06,
|
| 485 |
+
"loss": 0.6833,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.04416,
|
| 490 |
+
"grad_norm": 1.0460131168365479,
|
| 491 |
+
"learning_rate": 2.9400469383400897e-06,
|
| 492 |
+
"loss": 0.6802,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.0448,
|
| 497 |
+
"grad_norm": 1.2445200681686401,
|
| 498 |
+
"learning_rate": 2.9827181566033713e-06,
|
| 499 |
+
"loss": 0.6883,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.04544,
|
| 504 |
+
"grad_norm": 1.1759636402130127,
|
| 505 |
+
"learning_rate": 3.025389374866653e-06,
|
| 506 |
+
"loss": 0.682,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.04608,
|
| 511 |
+
"grad_norm": 1.2478561401367188,
|
| 512 |
+
"learning_rate": 3.068060593129934e-06,
|
| 513 |
+
"loss": 0.6802,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.04672,
|
| 518 |
+
"grad_norm": 0.8945108652114868,
|
| 519 |
+
"learning_rate": 3.1107318113932156e-06,
|
| 520 |
+
"loss": 0.6873,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.04736,
|
| 525 |
+
"grad_norm": 1.6810317039489746,
|
| 526 |
+
"learning_rate": 3.153403029656497e-06,
|
| 527 |
+
"loss": 0.6828,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.048,
|
| 532 |
+
"grad_norm": 2.882283926010132,
|
| 533 |
+
"learning_rate": 3.1960742479197783e-06,
|
| 534 |
+
"loss": 0.6797,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.04864,
|
| 539 |
+
"grad_norm": 1.575766921043396,
|
| 540 |
+
"learning_rate": 3.23874546618306e-06,
|
| 541 |
+
"loss": 0.6826,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.04928,
|
| 546 |
+
"grad_norm": 1.8044737577438354,
|
| 547 |
+
"learning_rate": 3.2814166844463415e-06,
|
| 548 |
+
"loss": 0.6837,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.04992,
|
| 553 |
+
"grad_norm": 1.4755513668060303,
|
| 554 |
+
"learning_rate": 3.324087902709623e-06,
|
| 555 |
+
"loss": 0.6815,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.05056,
|
| 560 |
+
"grad_norm": 2.0735654830932617,
|
| 561 |
+
"learning_rate": 3.3667591209729038e-06,
|
| 562 |
+
"loss": 0.6773,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.0512,
|
| 567 |
+
"grad_norm": 0.9823655486106873,
|
| 568 |
+
"learning_rate": 3.4094303392361854e-06,
|
| 569 |
+
"loss": 0.6689,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.05184,
|
| 574 |
+
"grad_norm": 1.0731920003890991,
|
| 575 |
+
"learning_rate": 3.452101557499467e-06,
|
| 576 |
+
"loss": 0.6689,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.05248,
|
| 581 |
+
"grad_norm": 1.1081087589263916,
|
| 582 |
+
"learning_rate": 3.494772775762748e-06,
|
| 583 |
+
"loss": 0.6694,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.05312,
|
| 588 |
+
"grad_norm": 1.2130361795425415,
|
| 589 |
+
"learning_rate": 3.5374439940260297e-06,
|
| 590 |
+
"loss": 0.682,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.05376,
|
| 595 |
+
"grad_norm": 1.170345664024353,
|
| 596 |
+
"learning_rate": 3.5801152122893113e-06,
|
| 597 |
+
"loss": 0.6756,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.0544,
|
| 602 |
+
"grad_norm": 1.2610124349594116,
|
| 603 |
+
"learning_rate": 3.622786430552593e-06,
|
| 604 |
+
"loss": 0.67,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.05504,
|
| 609 |
+
"grad_norm": 1.0138545036315918,
|
| 610 |
+
"learning_rate": 3.665457648815874e-06,
|
| 611 |
+
"loss": 0.6807,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.05568,
|
| 616 |
+
"grad_norm": 1.0937036275863647,
|
| 617 |
+
"learning_rate": 3.7081288670791556e-06,
|
| 618 |
+
"loss": 0.6774,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.05632,
|
| 623 |
+
"grad_norm": 1.119221568107605,
|
| 624 |
+
"learning_rate": 3.750800085342437e-06,
|
| 625 |
+
"loss": 0.6728,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.05696,
|
| 630 |
+
"grad_norm": 1.9351646900177002,
|
| 631 |
+
"learning_rate": 3.793471303605718e-06,
|
| 632 |
+
"loss": 0.6664,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.0576,
|
| 637 |
+
"grad_norm": 1.6208665370941162,
|
| 638 |
+
"learning_rate": 3.836142521869e-06,
|
| 639 |
+
"loss": 0.6653,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.05824,
|
| 644 |
+
"grad_norm": 1.3460793495178223,
|
| 645 |
+
"learning_rate": 3.8788137401322815e-06,
|
| 646 |
+
"loss": 0.6651,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.05888,
|
| 651 |
+
"grad_norm": 1.8987629413604736,
|
| 652 |
+
"learning_rate": 3.921484958395563e-06,
|
| 653 |
+
"loss": 0.6694,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.05952,
|
| 658 |
+
"grad_norm": 1.3637055158615112,
|
| 659 |
+
"learning_rate": 3.964156176658844e-06,
|
| 660 |
+
"loss": 0.6691,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.06016,
|
| 665 |
+
"grad_norm": 1.3319580554962158,
|
| 666 |
+
"learning_rate": 4.006827394922125e-06,
|
| 667 |
+
"loss": 0.6749,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.0608,
|
| 672 |
+
"grad_norm": 1.3574756383895874,
|
| 673 |
+
"learning_rate": 4.049498613185407e-06,
|
| 674 |
+
"loss": 0.6688,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.06144,
|
| 679 |
+
"grad_norm": 1.136319637298584,
|
| 680 |
+
"learning_rate": 4.092169831448688e-06,
|
| 681 |
+
"loss": 0.6657,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.06208,
|
| 686 |
+
"grad_norm": 1.7666376829147339,
|
| 687 |
+
"learning_rate": 4.134841049711969e-06,
|
| 688 |
+
"loss": 0.6667,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.06272,
|
| 693 |
+
"grad_norm": 1.332014560699463,
|
| 694 |
+
"learning_rate": 4.177512267975251e-06,
|
| 695 |
+
"loss": 0.6715,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.06336,
|
| 700 |
+
"grad_norm": 1.3127824068069458,
|
| 701 |
+
"learning_rate": 4.220183486238532e-06,
|
| 702 |
+
"loss": 0.6737,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.064,
|
| 707 |
+
"grad_norm": 1.293678641319275,
|
| 708 |
+
"learning_rate": 4.262854704501814e-06,
|
| 709 |
+
"loss": 0.6557,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.064,
|
| 714 |
+
"eval_accuracy": 0.76589,
|
| 715 |
+
"eval_loss": 0.6282070279121399,
|
| 716 |
+
"eval_macro_f1": 0.7657238614598489,
|
| 717 |
+
"eval_precision": 0.7708542319194018,
|
| 718 |
+
"eval_recall": 0.7687514011652705,
|
| 719 |
+
"eval_runtime": 806.5969,
|
| 720 |
+
"eval_samples_per_second": 123.978,
|
| 721 |
+
"eval_steps_per_second": 1.938,
|
| 722 |
+
"step": 1000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.06464,
|
| 726 |
+
"grad_norm": 1.2433074712753296,
|
| 727 |
+
"learning_rate": 4.3055259227650956e-06,
|
| 728 |
+
"loss": 0.6583,
|
| 729 |
+
"step": 1010
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.06528,
|
| 733 |
+
"grad_norm": 1.5115686655044556,
|
| 734 |
+
"learning_rate": 4.348197141028377e-06,
|
| 735 |
+
"loss": 0.6659,
|
| 736 |
+
"step": 1020
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.06592,
|
| 740 |
+
"grad_norm": 1.3560088872909546,
|
| 741 |
+
"learning_rate": 4.390868359291658e-06,
|
| 742 |
+
"loss": 0.6675,
|
| 743 |
+
"step": 1030
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.06656,
|
| 747 |
+
"grad_norm": 0.8229928016662598,
|
| 748 |
+
"learning_rate": 4.4335395775549394e-06,
|
| 749 |
+
"loss": 0.6629,
|
| 750 |
+
"step": 1040
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.0672,
|
| 754 |
+
"grad_norm": 1.5138262510299683,
|
| 755 |
+
"learning_rate": 4.476210795818221e-06,
|
| 756 |
+
"loss": 0.6635,
|
| 757 |
+
"step": 1050
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 0.06784,
|
| 761 |
+
"grad_norm": 1.1732233762741089,
|
| 762 |
+
"learning_rate": 4.518882014081503e-06,
|
| 763 |
+
"loss": 0.6531,
|
| 764 |
+
"step": 1060
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.06848,
|
| 768 |
+
"grad_norm": 1.6118066310882568,
|
| 769 |
+
"learning_rate": 4.561553232344783e-06,
|
| 770 |
+
"loss": 0.6644,
|
| 771 |
+
"step": 1070
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 0.06912,
|
| 775 |
+
"grad_norm": 0.9575800895690918,
|
| 776 |
+
"learning_rate": 4.604224450608065e-06,
|
| 777 |
+
"loss": 0.6542,
|
| 778 |
+
"step": 1080
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.06976,
|
| 782 |
+
"grad_norm": 1.157271146774292,
|
| 783 |
+
"learning_rate": 4.6468956688713465e-06,
|
| 784 |
+
"loss": 0.6505,
|
| 785 |
+
"step": 1090
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.0704,
|
| 789 |
+
"grad_norm": 1.4502589702606201,
|
| 790 |
+
"learning_rate": 4.689566887134628e-06,
|
| 791 |
+
"loss": 0.6725,
|
| 792 |
+
"step": 1100
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 0.07104,
|
| 796 |
+
"grad_norm": 1.122251033782959,
|
| 797 |
+
"learning_rate": 4.73223810539791e-06,
|
| 798 |
+
"loss": 0.6625,
|
| 799 |
+
"step": 1110
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 0.07168,
|
| 803 |
+
"grad_norm": 0.9529566764831543,
|
| 804 |
+
"learning_rate": 4.774909323661191e-06,
|
| 805 |
+
"loss": 0.6684,
|
| 806 |
+
"step": 1120
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.07232,
|
| 810 |
+
"grad_norm": 1.6433424949645996,
|
| 811 |
+
"learning_rate": 4.817580541924473e-06,
|
| 812 |
+
"loss": 0.6599,
|
| 813 |
+
"step": 1130
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.07296,
|
| 817 |
+
"grad_norm": 1.6565576791763306,
|
| 818 |
+
"learning_rate": 4.8602517601877535e-06,
|
| 819 |
+
"loss": 0.6591,
|
| 820 |
+
"step": 1140
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 0.0736,
|
| 824 |
+
"grad_norm": 1.3955042362213135,
|
| 825 |
+
"learning_rate": 4.902922978451035e-06,
|
| 826 |
+
"loss": 0.6649,
|
| 827 |
+
"step": 1150
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 0.07424,
|
| 831 |
+
"grad_norm": 1.530819296836853,
|
| 832 |
+
"learning_rate": 4.945594196714317e-06,
|
| 833 |
+
"loss": 0.65,
|
| 834 |
+
"step": 1160
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.07488,
|
| 838 |
+
"grad_norm": 1.0729092359542847,
|
| 839 |
+
"learning_rate": 4.9882654149775974e-06,
|
| 840 |
+
"loss": 0.6603,
|
| 841 |
+
"step": 1170
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.07552,
|
| 845 |
+
"grad_norm": 1.6307648420333862,
|
| 846 |
+
"learning_rate": 5.030936633240879e-06,
|
| 847 |
+
"loss": 0.6702,
|
| 848 |
+
"step": 1180
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.07616,
|
| 852 |
+
"grad_norm": 1.324406623840332,
|
| 853 |
+
"learning_rate": 5.073607851504161e-06,
|
| 854 |
+
"loss": 0.6634,
|
| 855 |
+
"step": 1190
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 0.0768,
|
| 859 |
+
"grad_norm": 1.1919140815734863,
|
| 860 |
+
"learning_rate": 5.116279069767442e-06,
|
| 861 |
+
"loss": 0.6586,
|
| 862 |
+
"step": 1200
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 0.07744,
|
| 866 |
+
"grad_norm": 1.0658751726150513,
|
| 867 |
+
"learning_rate": 5.158950288030724e-06,
|
| 868 |
+
"loss": 0.6573,
|
| 869 |
+
"step": 1210
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 0.07808,
|
| 873 |
+
"grad_norm": 1.7762951850891113,
|
| 874 |
+
"learning_rate": 5.201621506294005e-06,
|
| 875 |
+
"loss": 0.6596,
|
| 876 |
+
"step": 1220
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 0.07872,
|
| 880 |
+
"grad_norm": 1.1911667585372925,
|
| 881 |
+
"learning_rate": 5.244292724557286e-06,
|
| 882 |
+
"loss": 0.651,
|
| 883 |
+
"step": 1230
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 0.07936,
|
| 887 |
+
"grad_norm": 1.442029595375061,
|
| 888 |
+
"learning_rate": 5.286963942820568e-06,
|
| 889 |
+
"loss": 0.6598,
|
| 890 |
+
"step": 1240
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.08,
|
| 894 |
+
"grad_norm": 0.9955100417137146,
|
| 895 |
+
"learning_rate": 5.329635161083849e-06,
|
| 896 |
+
"loss": 0.6587,
|
| 897 |
+
"step": 1250
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.08064,
|
| 901 |
+
"grad_norm": 1.1801563501358032,
|
| 902 |
+
"learning_rate": 5.372306379347131e-06,
|
| 903 |
+
"loss": 0.6669,
|
| 904 |
+
"step": 1260
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.08128,
|
| 908 |
+
"grad_norm": 1.8548506498336792,
|
| 909 |
+
"learning_rate": 5.414977597610412e-06,
|
| 910 |
+
"loss": 0.6534,
|
| 911 |
+
"step": 1270
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.08192,
|
| 915 |
+
"grad_norm": 1.8297194242477417,
|
| 916 |
+
"learning_rate": 5.457648815873694e-06,
|
| 917 |
+
"loss": 0.668,
|
| 918 |
+
"step": 1280
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 0.08256,
|
| 922 |
+
"grad_norm": 1.3471113443374634,
|
| 923 |
+
"learning_rate": 5.5003200341369755e-06,
|
| 924 |
+
"loss": 0.659,
|
| 925 |
+
"step": 1290
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 0.0832,
|
| 929 |
+
"grad_norm": 1.4015250205993652,
|
| 930 |
+
"learning_rate": 5.542991252400256e-06,
|
| 931 |
+
"loss": 0.6602,
|
| 932 |
+
"step": 1300
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.08384,
|
| 936 |
+
"grad_norm": 1.0796362161636353,
|
| 937 |
+
"learning_rate": 5.585662470663538e-06,
|
| 938 |
+
"loss": 0.6652,
|
| 939 |
+
"step": 1310
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.08448,
|
| 943 |
+
"grad_norm": 2.0007293224334717,
|
| 944 |
+
"learning_rate": 5.628333688926819e-06,
|
| 945 |
+
"loss": 0.6573,
|
| 946 |
+
"step": 1320
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.08512,
|
| 950 |
+
"grad_norm": 1.3140546083450317,
|
| 951 |
+
"learning_rate": 5.671004907190101e-06,
|
| 952 |
+
"loss": 0.6642,
|
| 953 |
+
"step": 1330
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.08576,
|
| 957 |
+
"grad_norm": 1.6801965236663818,
|
| 958 |
+
"learning_rate": 5.7136761254533826e-06,
|
| 959 |
+
"loss": 0.6527,
|
| 960 |
+
"step": 1340
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.0864,
|
| 964 |
+
"grad_norm": 1.9289913177490234,
|
| 965 |
+
"learning_rate": 5.756347343716664e-06,
|
| 966 |
+
"loss": 0.6567,
|
| 967 |
+
"step": 1350
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 0.08704,
|
| 971 |
+
"grad_norm": 1.420436143875122,
|
| 972 |
+
"learning_rate": 5.799018561979946e-06,
|
| 973 |
+
"loss": 0.6479,
|
| 974 |
+
"step": 1360
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.08768,
|
| 978 |
+
"grad_norm": 1.3068914413452148,
|
| 979 |
+
"learning_rate": 5.841689780243226e-06,
|
| 980 |
+
"loss": 0.6522,
|
| 981 |
+
"step": 1370
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 0.08832,
|
| 985 |
+
"grad_norm": 1.8793392181396484,
|
| 986 |
+
"learning_rate": 5.884360998506507e-06,
|
| 987 |
+
"loss": 0.6491,
|
| 988 |
+
"step": 1380
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 0.08896,
|
| 992 |
+
"grad_norm": 1.3998247385025024,
|
| 993 |
+
"learning_rate": 5.927032216769789e-06,
|
| 994 |
+
"loss": 0.6638,
|
| 995 |
+
"step": 1390
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 0.0896,
|
| 999 |
+
"grad_norm": 1.371824026107788,
|
| 1000 |
+
"learning_rate": 5.96970343503307e-06,
|
| 1001 |
+
"loss": 0.6517,
|
| 1002 |
+
"step": 1400
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.09024,
|
| 1006 |
+
"grad_norm": 2.0050251483917236,
|
| 1007 |
+
"learning_rate": 6.012374653296352e-06,
|
| 1008 |
+
"loss": 0.6556,
|
| 1009 |
+
"step": 1410
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.09088,
|
| 1013 |
+
"grad_norm": 1.499855399131775,
|
| 1014 |
+
"learning_rate": 6.0550458715596335e-06,
|
| 1015 |
+
"loss": 0.6597,
|
| 1016 |
+
"step": 1420
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.09152,
|
| 1020 |
+
"grad_norm": 1.2066516876220703,
|
| 1021 |
+
"learning_rate": 6.097717089822915e-06,
|
| 1022 |
+
"loss": 0.6546,
|
| 1023 |
+
"step": 1430
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.09216,
|
| 1027 |
+
"grad_norm": 1.3586395978927612,
|
| 1028 |
+
"learning_rate": 6.140388308086196e-06,
|
| 1029 |
+
"loss": 0.6434,
|
| 1030 |
+
"step": 1440
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.0928,
|
| 1034 |
+
"grad_norm": 1.1413036584854126,
|
| 1035 |
+
"learning_rate": 6.183059526349477e-06,
|
| 1036 |
+
"loss": 0.6554,
|
| 1037 |
+
"step": 1450
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.09344,
|
| 1041 |
+
"grad_norm": 1.1844472885131836,
|
| 1042 |
+
"learning_rate": 6.225730744612759e-06,
|
| 1043 |
+
"loss": 0.6357,
|
| 1044 |
+
"step": 1460
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.09408,
|
| 1048 |
+
"grad_norm": 1.7203835248947144,
|
| 1049 |
+
"learning_rate": 6.2684019628760406e-06,
|
| 1050 |
+
"loss": 0.6546,
|
| 1051 |
+
"step": 1470
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.09472,
|
| 1055 |
+
"grad_norm": 1.1869184970855713,
|
| 1056 |
+
"learning_rate": 6.311073181139322e-06,
|
| 1057 |
+
"loss": 0.6376,
|
| 1058 |
+
"step": 1480
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.09536,
|
| 1062 |
+
"grad_norm": 1.3234835863113403,
|
| 1063 |
+
"learning_rate": 6.353744399402604e-06,
|
| 1064 |
+
"loss": 0.6494,
|
| 1065 |
+
"step": 1490
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.096,
|
| 1069 |
+
"grad_norm": 1.1924010515213013,
|
| 1070 |
+
"learning_rate": 6.396415617665885e-06,
|
| 1071 |
+
"loss": 0.652,
|
| 1072 |
+
"step": 1500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.09664,
|
| 1076 |
+
"grad_norm": 1.4505853652954102,
|
| 1077 |
+
"learning_rate": 6.439086835929166e-06,
|
| 1078 |
+
"loss": 0.6606,
|
| 1079 |
+
"step": 1510
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.09728,
|
| 1083 |
+
"grad_norm": 1.6351200342178345,
|
| 1084 |
+
"learning_rate": 6.481758054192448e-06,
|
| 1085 |
+
"loss": 0.6567,
|
| 1086 |
+
"step": 1520
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.09792,
|
| 1090 |
+
"grad_norm": 1.4423909187316895,
|
| 1091 |
+
"learning_rate": 6.524429272455729e-06,
|
| 1092 |
+
"loss": 0.653,
|
| 1093 |
+
"step": 1530
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 0.09856,
|
| 1097 |
+
"grad_norm": 0.8749169111251831,
|
| 1098 |
+
"learning_rate": 6.567100490719011e-06,
|
| 1099 |
+
"loss": 0.6417,
|
| 1100 |
+
"step": 1540
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 0.0992,
|
| 1104 |
+
"grad_norm": 1.2707208395004272,
|
| 1105 |
+
"learning_rate": 6.609771708982292e-06,
|
| 1106 |
+
"loss": 0.6563,
|
| 1107 |
+
"step": 1550
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 0.09984,
|
| 1111 |
+
"grad_norm": 1.163254976272583,
|
| 1112 |
+
"learning_rate": 6.652442927245574e-06,
|
| 1113 |
+
"loss": 0.6425,
|
| 1114 |
+
"step": 1560
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 0.10048,
|
| 1118 |
+
"grad_norm": 1.4674128293991089,
|
| 1119 |
+
"learning_rate": 6.6951141455088555e-06,
|
| 1120 |
+
"loss": 0.6561,
|
| 1121 |
+
"step": 1570
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 0.10112,
|
| 1125 |
+
"grad_norm": 1.6261919736862183,
|
| 1126 |
+
"learning_rate": 6.737785363772135e-06,
|
| 1127 |
+
"loss": 0.6473,
|
| 1128 |
+
"step": 1580
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 0.10176,
|
| 1132 |
+
"grad_norm": 1.2146090269088745,
|
| 1133 |
+
"learning_rate": 6.780456582035417e-06,
|
| 1134 |
+
"loss": 0.6467,
|
| 1135 |
+
"step": 1590
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 0.1024,
|
| 1139 |
+
"grad_norm": 1.3596712350845337,
|
| 1140 |
+
"learning_rate": 6.8231278002986985e-06,
|
| 1141 |
+
"loss": 0.6431,
|
| 1142 |
+
"step": 1600
|
| 1143 |
+
}
|
| 1144 |
+
],
|
| 1145 |
+
"logging_steps": 10,
|
| 1146 |
+
"max_steps": 46875,
|
| 1147 |
+
"num_input_tokens_seen": 0,
|
| 1148 |
+
"num_train_epochs": 3,
|
| 1149 |
+
"save_steps": 200,
|
| 1150 |
+
"stateful_callbacks": {
|
| 1151 |
+
"EarlyStoppingCallback": {
|
| 1152 |
+
"args": {
|
| 1153 |
+
"early_stopping_patience": 3,
|
| 1154 |
+
"early_stopping_threshold": 0.0
|
| 1155 |
+
},
|
| 1156 |
+
"attributes": {
|
| 1157 |
+
"early_stopping_patience_counter": 0
|
| 1158 |
+
}
|
| 1159 |
+
},
|
| 1160 |
+
"TrainerControl": {
|
| 1161 |
+
"args": {
|
| 1162 |
+
"should_epoch_stop": false,
|
| 1163 |
+
"should_evaluate": false,
|
| 1164 |
+
"should_log": false,
|
| 1165 |
+
"should_save": true,
|
| 1166 |
+
"should_training_stop": false
|
| 1167 |
+
},
|
| 1168 |
+
"attributes": {}
|
| 1169 |
+
}
|
| 1170 |
+
},
|
| 1171 |
+
"total_flos": 1.346581003559232e+16,
|
| 1172 |
+
"train_batch_size": 32,
|
| 1173 |
+
"trial_name": null,
|
| 1174 |
+
"trial_params": null
|
| 1175 |
+
}
|
graphcodebert-rdrop/checkpoint-1600/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a0e6c50c4bb64305fe23a4e428084f9c072417915284e4ed838c0b852405eb80
|
| 3 |
+
size 5841
|
graphcodebert-rdrop/checkpoint-1600/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
graphcodebert-rdrop/checkpoint-1800/config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.2,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"gradient_checkpointing": false,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.2,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 514,
|
| 18 |
+
"model_type": "roberta",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"output_past": true,
|
| 22 |
+
"pad_token_id": 1,
|
| 23 |
+
"position_embedding_type": "absolute",
|
| 24 |
+
"transformers_version": "4.56.0",
|
| 25 |
+
"type_vocab_size": 1,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 50265
|
| 28 |
+
}
|
graphcodebert-rdrop/checkpoint-1800/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
graphcodebert-rdrop/checkpoint-1800/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51808bd23f6f3b2694c4effdb372be4beda8d4cd21d7e03a6ce4216bad7a01e7
|
| 3 |
+
size 498612824
|
graphcodebert-rdrop/checkpoint-1800/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:88fdf16b62a54ab79547842b5da1a1d264b9e61aa634f3b5182194dadabc4b0e
|
| 3 |
+
size 4741923
|
graphcodebert-rdrop/checkpoint-1800/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6175af2948633572f388d48ee84616b36fde290ef723a7f0176fed3f59123429
|
| 3 |
+
size 14645
|
graphcodebert-rdrop/checkpoint-1800/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa0408efb69cab96d5bab9a1aaf44cedbc9fc8d34f4cef378d81605e5c026d5c
|
| 3 |
+
size 1383
|
graphcodebert-rdrop/checkpoint-1800/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f42b82fc050daf3e912abc97eef91deaf065d469a1aa16fd4898dd0826aa8836
|
| 3 |
+
size 1465
|
graphcodebert-rdrop/checkpoint-1800/special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
graphcodebert-rdrop/checkpoint-1800/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
graphcodebert-rdrop/checkpoint-1800/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 512,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 56 |
+
"trim_offsets": true,
|
| 57 |
+
"unk_token": "<unk>"
|
| 58 |
+
}
|
graphcodebert-rdrop/checkpoint-1800/trainer_state.json
ADDED
|
@@ -0,0 +1,1315 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1000,
|
| 3 |
+
"best_metric": 0.7657238614598489,
|
| 4 |
+
"best_model_checkpoint": "./output_checkpoints/graphcodebert-rdrop/checkpoint-1000",
|
| 5 |
+
"epoch": 0.1152,
|
| 6 |
+
"eval_steps": 1000,
|
| 7 |
+
"global_step": 1800,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00064,
|
| 14 |
+
"grad_norm": 1.8688431978225708,
|
| 15 |
+
"learning_rate": 3.840409643695328e-08,
|
| 16 |
+
"loss": 0.7168,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.00128,
|
| 21 |
+
"grad_norm": 1.835353970527649,
|
| 22 |
+
"learning_rate": 8.10753147002347e-08,
|
| 23 |
+
"loss": 0.7179,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00192,
|
| 28 |
+
"grad_norm": 1.2541388273239136,
|
| 29 |
+
"learning_rate": 1.2374653296351612e-07,
|
| 30 |
+
"loss": 0.7177,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.00256,
|
| 35 |
+
"grad_norm": 1.1381633281707764,
|
| 36 |
+
"learning_rate": 1.6641775122679754e-07,
|
| 37 |
+
"loss": 0.7312,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0032,
|
| 42 |
+
"grad_norm": 1.6313824653625488,
|
| 43 |
+
"learning_rate": 2.0908896949007894e-07,
|
| 44 |
+
"loss": 0.7167,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.00384,
|
| 49 |
+
"grad_norm": 1.092464566230774,
|
| 50 |
+
"learning_rate": 2.517601877533604e-07,
|
| 51 |
+
"loss": 0.7201,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.00448,
|
| 56 |
+
"grad_norm": 1.164099097251892,
|
| 57 |
+
"learning_rate": 2.944314060166418e-07,
|
| 58 |
+
"loss": 0.7195,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.00512,
|
| 63 |
+
"grad_norm": 1.9283920526504517,
|
| 64 |
+
"learning_rate": 3.371026242799232e-07,
|
| 65 |
+
"loss": 0.7223,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.00576,
|
| 70 |
+
"grad_norm": 1.8716129064559937,
|
| 71 |
+
"learning_rate": 3.7977384254320464e-07,
|
| 72 |
+
"loss": 0.7151,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0064,
|
| 77 |
+
"grad_norm": 1.5499262809753418,
|
| 78 |
+
"learning_rate": 4.22445060806486e-07,
|
| 79 |
+
"loss": 0.7211,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.00704,
|
| 84 |
+
"grad_norm": 1.375114917755127,
|
| 85 |
+
"learning_rate": 4.651162790697675e-07,
|
| 86 |
+
"loss": 0.7247,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.00768,
|
| 91 |
+
"grad_norm": 1.384915828704834,
|
| 92 |
+
"learning_rate": 5.077874973330489e-07,
|
| 93 |
+
"loss": 0.7156,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.00832,
|
| 98 |
+
"grad_norm": 1.6376659870147705,
|
| 99 |
+
"learning_rate": 5.504587155963304e-07,
|
| 100 |
+
"loss": 0.7196,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.00896,
|
| 105 |
+
"grad_norm": 1.709489345550537,
|
| 106 |
+
"learning_rate": 5.931299338596117e-07,
|
| 107 |
+
"loss": 0.7215,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0096,
|
| 112 |
+
"grad_norm": 1.350512146949768,
|
| 113 |
+
"learning_rate": 6.358011521228932e-07,
|
| 114 |
+
"loss": 0.71,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.01024,
|
| 119 |
+
"grad_norm": 2.33050537109375,
|
| 120 |
+
"learning_rate": 6.784723703861745e-07,
|
| 121 |
+
"loss": 0.7191,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.01088,
|
| 126 |
+
"grad_norm": 1.0042874813079834,
|
| 127 |
+
"learning_rate": 7.21143588649456e-07,
|
| 128 |
+
"loss": 0.72,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.01152,
|
| 133 |
+
"grad_norm": 1.1835744380950928,
|
| 134 |
+
"learning_rate": 7.638148069127374e-07,
|
| 135 |
+
"loss": 0.7122,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.01216,
|
| 140 |
+
"grad_norm": 1.949506402015686,
|
| 141 |
+
"learning_rate": 8.064860251760189e-07,
|
| 142 |
+
"loss": 0.7091,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.0128,
|
| 147 |
+
"grad_norm": 1.139626383781433,
|
| 148 |
+
"learning_rate": 8.491572434393003e-07,
|
| 149 |
+
"loss": 0.7046,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.01344,
|
| 154 |
+
"grad_norm": 1.7734779119491577,
|
| 155 |
+
"learning_rate": 8.918284617025817e-07,
|
| 156 |
+
"loss": 0.7135,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.01408,
|
| 161 |
+
"grad_norm": 1.3442974090576172,
|
| 162 |
+
"learning_rate": 9.344996799658632e-07,
|
| 163 |
+
"loss": 0.7127,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.01472,
|
| 168 |
+
"grad_norm": 1.6148335933685303,
|
| 169 |
+
"learning_rate": 9.771708982291445e-07,
|
| 170 |
+
"loss": 0.7168,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.01536,
|
| 175 |
+
"grad_norm": 0.9220213294029236,
|
| 176 |
+
"learning_rate": 1.0198421164924258e-06,
|
| 177 |
+
"loss": 0.7109,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.016,
|
| 182 |
+
"grad_norm": 1.5631815195083618,
|
| 183 |
+
"learning_rate": 1.0625133347557074e-06,
|
| 184 |
+
"loss": 0.7149,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.01664,
|
| 189 |
+
"grad_norm": 1.5052822828292847,
|
| 190 |
+
"learning_rate": 1.1051845530189888e-06,
|
| 191 |
+
"loss": 0.7112,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.01728,
|
| 196 |
+
"grad_norm": 1.6714903116226196,
|
| 197 |
+
"learning_rate": 1.1478557712822702e-06,
|
| 198 |
+
"loss": 0.7128,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.01792,
|
| 203 |
+
"grad_norm": 1.6667262315750122,
|
| 204 |
+
"learning_rate": 1.1905269895455517e-06,
|
| 205 |
+
"loss": 0.7091,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.01856,
|
| 210 |
+
"grad_norm": 1.6861907243728638,
|
| 211 |
+
"learning_rate": 1.233198207808833e-06,
|
| 212 |
+
"loss": 0.7078,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0192,
|
| 217 |
+
"grad_norm": 1.6259212493896484,
|
| 218 |
+
"learning_rate": 1.2758694260721145e-06,
|
| 219 |
+
"loss": 0.7087,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.01984,
|
| 224 |
+
"grad_norm": 1.207320213317871,
|
| 225 |
+
"learning_rate": 1.318540644335396e-06,
|
| 226 |
+
"loss": 0.7087,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.02048,
|
| 231 |
+
"grad_norm": 1.087124228477478,
|
| 232 |
+
"learning_rate": 1.3612118625986772e-06,
|
| 233 |
+
"loss": 0.7174,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.02112,
|
| 238 |
+
"grad_norm": 1.5668721199035645,
|
| 239 |
+
"learning_rate": 1.4038830808619588e-06,
|
| 240 |
+
"loss": 0.7066,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.02176,
|
| 245 |
+
"grad_norm": 1.5332859754562378,
|
| 246 |
+
"learning_rate": 1.4465542991252401e-06,
|
| 247 |
+
"loss": 0.7115,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0224,
|
| 252 |
+
"grad_norm": 2.092994451522827,
|
| 253 |
+
"learning_rate": 1.4892255173885215e-06,
|
| 254 |
+
"loss": 0.7085,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.02304,
|
| 259 |
+
"grad_norm": 1.266851544380188,
|
| 260 |
+
"learning_rate": 1.531896735651803e-06,
|
| 261 |
+
"loss": 0.7091,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.02368,
|
| 266 |
+
"grad_norm": 1.290616512298584,
|
| 267 |
+
"learning_rate": 1.5745679539150842e-06,
|
| 268 |
+
"loss": 0.7103,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.02432,
|
| 273 |
+
"grad_norm": 1.2045000791549683,
|
| 274 |
+
"learning_rate": 1.6172391721783658e-06,
|
| 275 |
+
"loss": 0.705,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.02496,
|
| 280 |
+
"grad_norm": 1.828832983970642,
|
| 281 |
+
"learning_rate": 1.6599103904416472e-06,
|
| 282 |
+
"loss": 0.7038,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.0256,
|
| 287 |
+
"grad_norm": 1.7793196439743042,
|
| 288 |
+
"learning_rate": 1.7025816087049288e-06,
|
| 289 |
+
"loss": 0.7085,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.02624,
|
| 294 |
+
"grad_norm": 0.8949472904205322,
|
| 295 |
+
"learning_rate": 1.7452528269682101e-06,
|
| 296 |
+
"loss": 0.7045,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.02688,
|
| 301 |
+
"grad_norm": 1.3071945905685425,
|
| 302 |
+
"learning_rate": 1.7879240452314913e-06,
|
| 303 |
+
"loss": 0.702,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.02752,
|
| 308 |
+
"grad_norm": 1.7497148513793945,
|
| 309 |
+
"learning_rate": 1.8305952634947729e-06,
|
| 310 |
+
"loss": 0.7073,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.02816,
|
| 315 |
+
"grad_norm": 1.0856297016143799,
|
| 316 |
+
"learning_rate": 1.8732664817580542e-06,
|
| 317 |
+
"loss": 0.7008,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.0288,
|
| 322 |
+
"grad_norm": 1.369019627571106,
|
| 323 |
+
"learning_rate": 1.915937700021336e-06,
|
| 324 |
+
"loss": 0.7059,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.02944,
|
| 329 |
+
"grad_norm": 2.0961010456085205,
|
| 330 |
+
"learning_rate": 1.958608918284617e-06,
|
| 331 |
+
"loss": 0.6952,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.03008,
|
| 336 |
+
"grad_norm": 1.6076347827911377,
|
| 337 |
+
"learning_rate": 2.0012801365478988e-06,
|
| 338 |
+
"loss": 0.7074,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.03072,
|
| 343 |
+
"grad_norm": 1.3129311800003052,
|
| 344 |
+
"learning_rate": 2.04395135481118e-06,
|
| 345 |
+
"loss": 0.6994,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.03136,
|
| 350 |
+
"grad_norm": 1.3621476888656616,
|
| 351 |
+
"learning_rate": 2.0866225730744615e-06,
|
| 352 |
+
"loss": 0.696,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.032,
|
| 357 |
+
"grad_norm": 0.9786806106567383,
|
| 358 |
+
"learning_rate": 2.129293791337743e-06,
|
| 359 |
+
"loss": 0.7011,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.03264,
|
| 364 |
+
"grad_norm": 1.4611176252365112,
|
| 365 |
+
"learning_rate": 2.1719650096010242e-06,
|
| 366 |
+
"loss": 0.699,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.03328,
|
| 371 |
+
"grad_norm": 1.0675945281982422,
|
| 372 |
+
"learning_rate": 2.214636227864306e-06,
|
| 373 |
+
"loss": 0.6906,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.03392,
|
| 378 |
+
"grad_norm": 1.6564017534255981,
|
| 379 |
+
"learning_rate": 2.257307446127587e-06,
|
| 380 |
+
"loss": 0.6923,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.03456,
|
| 385 |
+
"grad_norm": 1.228119134902954,
|
| 386 |
+
"learning_rate": 2.2999786643908685e-06,
|
| 387 |
+
"loss": 0.6928,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.0352,
|
| 392 |
+
"grad_norm": 1.9868593215942383,
|
| 393 |
+
"learning_rate": 2.34264988265415e-06,
|
| 394 |
+
"loss": 0.6912,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.03584,
|
| 399 |
+
"grad_norm": 1.2531176805496216,
|
| 400 |
+
"learning_rate": 2.3853211009174317e-06,
|
| 401 |
+
"loss": 0.6923,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.03648,
|
| 406 |
+
"grad_norm": 1.413602352142334,
|
| 407 |
+
"learning_rate": 2.427992319180713e-06,
|
| 408 |
+
"loss": 0.6993,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.03712,
|
| 413 |
+
"grad_norm": 0.9655390977859497,
|
| 414 |
+
"learning_rate": 2.470663537443994e-06,
|
| 415 |
+
"loss": 0.7017,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.03776,
|
| 420 |
+
"grad_norm": 1.962438941001892,
|
| 421 |
+
"learning_rate": 2.5133347557072756e-06,
|
| 422 |
+
"loss": 0.6933,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.0384,
|
| 427 |
+
"grad_norm": 1.1099931001663208,
|
| 428 |
+
"learning_rate": 2.556005973970557e-06,
|
| 429 |
+
"loss": 0.6925,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.03904,
|
| 434 |
+
"grad_norm": 1.1766624450683594,
|
| 435 |
+
"learning_rate": 2.5986771922338383e-06,
|
| 436 |
+
"loss": 0.6894,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.03968,
|
| 441 |
+
"grad_norm": 1.574353814125061,
|
| 442 |
+
"learning_rate": 2.64134841049712e-06,
|
| 443 |
+
"loss": 0.6923,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.04032,
|
| 448 |
+
"grad_norm": 1.194074273109436,
|
| 449 |
+
"learning_rate": 2.6840196287604015e-06,
|
| 450 |
+
"loss": 0.6855,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.04096,
|
| 455 |
+
"grad_norm": 1.1750593185424805,
|
| 456 |
+
"learning_rate": 2.7266908470236826e-06,
|
| 457 |
+
"loss": 0.7002,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.0416,
|
| 462 |
+
"grad_norm": 1.3165347576141357,
|
| 463 |
+
"learning_rate": 2.7693620652869642e-06,
|
| 464 |
+
"loss": 0.6893,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.04224,
|
| 469 |
+
"grad_norm": 1.1079384088516235,
|
| 470 |
+
"learning_rate": 2.812033283550246e-06,
|
| 471 |
+
"loss": 0.6852,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.04288,
|
| 476 |
+
"grad_norm": 1.231327772140503,
|
| 477 |
+
"learning_rate": 2.8547045018135274e-06,
|
| 478 |
+
"loss": 0.695,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.04352,
|
| 483 |
+
"grad_norm": 1.966036319732666,
|
| 484 |
+
"learning_rate": 2.897375720076808e-06,
|
| 485 |
+
"loss": 0.6833,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.04416,
|
| 490 |
+
"grad_norm": 1.0460131168365479,
|
| 491 |
+
"learning_rate": 2.9400469383400897e-06,
|
| 492 |
+
"loss": 0.6802,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.0448,
|
| 497 |
+
"grad_norm": 1.2445200681686401,
|
| 498 |
+
"learning_rate": 2.9827181566033713e-06,
|
| 499 |
+
"loss": 0.6883,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.04544,
|
| 504 |
+
"grad_norm": 1.1759636402130127,
|
| 505 |
+
"learning_rate": 3.025389374866653e-06,
|
| 506 |
+
"loss": 0.682,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.04608,
|
| 511 |
+
"grad_norm": 1.2478561401367188,
|
| 512 |
+
"learning_rate": 3.068060593129934e-06,
|
| 513 |
+
"loss": 0.6802,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.04672,
|
| 518 |
+
"grad_norm": 0.8945108652114868,
|
| 519 |
+
"learning_rate": 3.1107318113932156e-06,
|
| 520 |
+
"loss": 0.6873,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.04736,
|
| 525 |
+
"grad_norm": 1.6810317039489746,
|
| 526 |
+
"learning_rate": 3.153403029656497e-06,
|
| 527 |
+
"loss": 0.6828,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.048,
|
| 532 |
+
"grad_norm": 2.882283926010132,
|
| 533 |
+
"learning_rate": 3.1960742479197783e-06,
|
| 534 |
+
"loss": 0.6797,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.04864,
|
| 539 |
+
"grad_norm": 1.575766921043396,
|
| 540 |
+
"learning_rate": 3.23874546618306e-06,
|
| 541 |
+
"loss": 0.6826,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.04928,
|
| 546 |
+
"grad_norm": 1.8044737577438354,
|
| 547 |
+
"learning_rate": 3.2814166844463415e-06,
|
| 548 |
+
"loss": 0.6837,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.04992,
|
| 553 |
+
"grad_norm": 1.4755513668060303,
|
| 554 |
+
"learning_rate": 3.324087902709623e-06,
|
| 555 |
+
"loss": 0.6815,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.05056,
|
| 560 |
+
"grad_norm": 2.0735654830932617,
|
| 561 |
+
"learning_rate": 3.3667591209729038e-06,
|
| 562 |
+
"loss": 0.6773,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.0512,
|
| 567 |
+
"grad_norm": 0.9823655486106873,
|
| 568 |
+
"learning_rate": 3.4094303392361854e-06,
|
| 569 |
+
"loss": 0.6689,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.05184,
|
| 574 |
+
"grad_norm": 1.0731920003890991,
|
| 575 |
+
"learning_rate": 3.452101557499467e-06,
|
| 576 |
+
"loss": 0.6689,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.05248,
|
| 581 |
+
"grad_norm": 1.1081087589263916,
|
| 582 |
+
"learning_rate": 3.494772775762748e-06,
|
| 583 |
+
"loss": 0.6694,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.05312,
|
| 588 |
+
"grad_norm": 1.2130361795425415,
|
| 589 |
+
"learning_rate": 3.5374439940260297e-06,
|
| 590 |
+
"loss": 0.682,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.05376,
|
| 595 |
+
"grad_norm": 1.170345664024353,
|
| 596 |
+
"learning_rate": 3.5801152122893113e-06,
|
| 597 |
+
"loss": 0.6756,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.0544,
|
| 602 |
+
"grad_norm": 1.2610124349594116,
|
| 603 |
+
"learning_rate": 3.622786430552593e-06,
|
| 604 |
+
"loss": 0.67,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.05504,
|
| 609 |
+
"grad_norm": 1.0138545036315918,
|
| 610 |
+
"learning_rate": 3.665457648815874e-06,
|
| 611 |
+
"loss": 0.6807,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.05568,
|
| 616 |
+
"grad_norm": 1.0937036275863647,
|
| 617 |
+
"learning_rate": 3.7081288670791556e-06,
|
| 618 |
+
"loss": 0.6774,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.05632,
|
| 623 |
+
"grad_norm": 1.119221568107605,
|
| 624 |
+
"learning_rate": 3.750800085342437e-06,
|
| 625 |
+
"loss": 0.6728,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.05696,
|
| 630 |
+
"grad_norm": 1.9351646900177002,
|
| 631 |
+
"learning_rate": 3.793471303605718e-06,
|
| 632 |
+
"loss": 0.6664,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.0576,
|
| 637 |
+
"grad_norm": 1.6208665370941162,
|
| 638 |
+
"learning_rate": 3.836142521869e-06,
|
| 639 |
+
"loss": 0.6653,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.05824,
|
| 644 |
+
"grad_norm": 1.3460793495178223,
|
| 645 |
+
"learning_rate": 3.8788137401322815e-06,
|
| 646 |
+
"loss": 0.6651,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.05888,
|
| 651 |
+
"grad_norm": 1.8987629413604736,
|
| 652 |
+
"learning_rate": 3.921484958395563e-06,
|
| 653 |
+
"loss": 0.6694,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.05952,
|
| 658 |
+
"grad_norm": 1.3637055158615112,
|
| 659 |
+
"learning_rate": 3.964156176658844e-06,
|
| 660 |
+
"loss": 0.6691,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.06016,
|
| 665 |
+
"grad_norm": 1.3319580554962158,
|
| 666 |
+
"learning_rate": 4.006827394922125e-06,
|
| 667 |
+
"loss": 0.6749,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.0608,
|
| 672 |
+
"grad_norm": 1.3574756383895874,
|
| 673 |
+
"learning_rate": 4.049498613185407e-06,
|
| 674 |
+
"loss": 0.6688,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.06144,
|
| 679 |
+
"grad_norm": 1.136319637298584,
|
| 680 |
+
"learning_rate": 4.092169831448688e-06,
|
| 681 |
+
"loss": 0.6657,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.06208,
|
| 686 |
+
"grad_norm": 1.7666376829147339,
|
| 687 |
+
"learning_rate": 4.134841049711969e-06,
|
| 688 |
+
"loss": 0.6667,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.06272,
|
| 693 |
+
"grad_norm": 1.332014560699463,
|
| 694 |
+
"learning_rate": 4.177512267975251e-06,
|
| 695 |
+
"loss": 0.6715,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.06336,
|
| 700 |
+
"grad_norm": 1.3127824068069458,
|
| 701 |
+
"learning_rate": 4.220183486238532e-06,
|
| 702 |
+
"loss": 0.6737,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.064,
|
| 707 |
+
"grad_norm": 1.293678641319275,
|
| 708 |
+
"learning_rate": 4.262854704501814e-06,
|
| 709 |
+
"loss": 0.6557,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.064,
|
| 714 |
+
"eval_accuracy": 0.76589,
|
| 715 |
+
"eval_loss": 0.6282070279121399,
|
| 716 |
+
"eval_macro_f1": 0.7657238614598489,
|
| 717 |
+
"eval_precision": 0.7708542319194018,
|
| 718 |
+
"eval_recall": 0.7687514011652705,
|
| 719 |
+
"eval_runtime": 806.5969,
|
| 720 |
+
"eval_samples_per_second": 123.978,
|
| 721 |
+
"eval_steps_per_second": 1.938,
|
| 722 |
+
"step": 1000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.06464,
|
| 726 |
+
"grad_norm": 1.2433074712753296,
|
| 727 |
+
"learning_rate": 4.3055259227650956e-06,
|
| 728 |
+
"loss": 0.6583,
|
| 729 |
+
"step": 1010
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.06528,
|
| 733 |
+
"grad_norm": 1.5115686655044556,
|
| 734 |
+
"learning_rate": 4.348197141028377e-06,
|
| 735 |
+
"loss": 0.6659,
|
| 736 |
+
"step": 1020
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.06592,
|
| 740 |
+
"grad_norm": 1.3560088872909546,
|
| 741 |
+
"learning_rate": 4.390868359291658e-06,
|
| 742 |
+
"loss": 0.6675,
|
| 743 |
+
"step": 1030
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.06656,
|
| 747 |
+
"grad_norm": 0.8229928016662598,
|
| 748 |
+
"learning_rate": 4.4335395775549394e-06,
|
| 749 |
+
"loss": 0.6629,
|
| 750 |
+
"step": 1040
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.0672,
|
| 754 |
+
"grad_norm": 1.5138262510299683,
|
| 755 |
+
"learning_rate": 4.476210795818221e-06,
|
| 756 |
+
"loss": 0.6635,
|
| 757 |
+
"step": 1050
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 0.06784,
|
| 761 |
+
"grad_norm": 1.1732233762741089,
|
| 762 |
+
"learning_rate": 4.518882014081503e-06,
|
| 763 |
+
"loss": 0.6531,
|
| 764 |
+
"step": 1060
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.06848,
|
| 768 |
+
"grad_norm": 1.6118066310882568,
|
| 769 |
+
"learning_rate": 4.561553232344783e-06,
|
| 770 |
+
"loss": 0.6644,
|
| 771 |
+
"step": 1070
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 0.06912,
|
| 775 |
+
"grad_norm": 0.9575800895690918,
|
| 776 |
+
"learning_rate": 4.604224450608065e-06,
|
| 777 |
+
"loss": 0.6542,
|
| 778 |
+
"step": 1080
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.06976,
|
| 782 |
+
"grad_norm": 1.157271146774292,
|
| 783 |
+
"learning_rate": 4.6468956688713465e-06,
|
| 784 |
+
"loss": 0.6505,
|
| 785 |
+
"step": 1090
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.0704,
|
| 789 |
+
"grad_norm": 1.4502589702606201,
|
| 790 |
+
"learning_rate": 4.689566887134628e-06,
|
| 791 |
+
"loss": 0.6725,
|
| 792 |
+
"step": 1100
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 0.07104,
|
| 796 |
+
"grad_norm": 1.122251033782959,
|
| 797 |
+
"learning_rate": 4.73223810539791e-06,
|
| 798 |
+
"loss": 0.6625,
|
| 799 |
+
"step": 1110
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 0.07168,
|
| 803 |
+
"grad_norm": 0.9529566764831543,
|
| 804 |
+
"learning_rate": 4.774909323661191e-06,
|
| 805 |
+
"loss": 0.6684,
|
| 806 |
+
"step": 1120
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.07232,
|
| 810 |
+
"grad_norm": 1.6433424949645996,
|
| 811 |
+
"learning_rate": 4.817580541924473e-06,
|
| 812 |
+
"loss": 0.6599,
|
| 813 |
+
"step": 1130
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.07296,
|
| 817 |
+
"grad_norm": 1.6565576791763306,
|
| 818 |
+
"learning_rate": 4.8602517601877535e-06,
|
| 819 |
+
"loss": 0.6591,
|
| 820 |
+
"step": 1140
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 0.0736,
|
| 824 |
+
"grad_norm": 1.3955042362213135,
|
| 825 |
+
"learning_rate": 4.902922978451035e-06,
|
| 826 |
+
"loss": 0.6649,
|
| 827 |
+
"step": 1150
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 0.07424,
|
| 831 |
+
"grad_norm": 1.530819296836853,
|
| 832 |
+
"learning_rate": 4.945594196714317e-06,
|
| 833 |
+
"loss": 0.65,
|
| 834 |
+
"step": 1160
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.07488,
|
| 838 |
+
"grad_norm": 1.0729092359542847,
|
| 839 |
+
"learning_rate": 4.9882654149775974e-06,
|
| 840 |
+
"loss": 0.6603,
|
| 841 |
+
"step": 1170
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.07552,
|
| 845 |
+
"grad_norm": 1.6307648420333862,
|
| 846 |
+
"learning_rate": 5.030936633240879e-06,
|
| 847 |
+
"loss": 0.6702,
|
| 848 |
+
"step": 1180
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.07616,
|
| 852 |
+
"grad_norm": 1.324406623840332,
|
| 853 |
+
"learning_rate": 5.073607851504161e-06,
|
| 854 |
+
"loss": 0.6634,
|
| 855 |
+
"step": 1190
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 0.0768,
|
| 859 |
+
"grad_norm": 1.1919140815734863,
|
| 860 |
+
"learning_rate": 5.116279069767442e-06,
|
| 861 |
+
"loss": 0.6586,
|
| 862 |
+
"step": 1200
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 0.07744,
|
| 866 |
+
"grad_norm": 1.0658751726150513,
|
| 867 |
+
"learning_rate": 5.158950288030724e-06,
|
| 868 |
+
"loss": 0.6573,
|
| 869 |
+
"step": 1210
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 0.07808,
|
| 873 |
+
"grad_norm": 1.7762951850891113,
|
| 874 |
+
"learning_rate": 5.201621506294005e-06,
|
| 875 |
+
"loss": 0.6596,
|
| 876 |
+
"step": 1220
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 0.07872,
|
| 880 |
+
"grad_norm": 1.1911667585372925,
|
| 881 |
+
"learning_rate": 5.244292724557286e-06,
|
| 882 |
+
"loss": 0.651,
|
| 883 |
+
"step": 1230
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 0.07936,
|
| 887 |
+
"grad_norm": 1.442029595375061,
|
| 888 |
+
"learning_rate": 5.286963942820568e-06,
|
| 889 |
+
"loss": 0.6598,
|
| 890 |
+
"step": 1240
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.08,
|
| 894 |
+
"grad_norm": 0.9955100417137146,
|
| 895 |
+
"learning_rate": 5.329635161083849e-06,
|
| 896 |
+
"loss": 0.6587,
|
| 897 |
+
"step": 1250
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.08064,
|
| 901 |
+
"grad_norm": 1.1801563501358032,
|
| 902 |
+
"learning_rate": 5.372306379347131e-06,
|
| 903 |
+
"loss": 0.6669,
|
| 904 |
+
"step": 1260
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.08128,
|
| 908 |
+
"grad_norm": 1.8548506498336792,
|
| 909 |
+
"learning_rate": 5.414977597610412e-06,
|
| 910 |
+
"loss": 0.6534,
|
| 911 |
+
"step": 1270
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.08192,
|
| 915 |
+
"grad_norm": 1.8297194242477417,
|
| 916 |
+
"learning_rate": 5.457648815873694e-06,
|
| 917 |
+
"loss": 0.668,
|
| 918 |
+
"step": 1280
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 0.08256,
|
| 922 |
+
"grad_norm": 1.3471113443374634,
|
| 923 |
+
"learning_rate": 5.5003200341369755e-06,
|
| 924 |
+
"loss": 0.659,
|
| 925 |
+
"step": 1290
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 0.0832,
|
| 929 |
+
"grad_norm": 1.4015250205993652,
|
| 930 |
+
"learning_rate": 5.542991252400256e-06,
|
| 931 |
+
"loss": 0.6602,
|
| 932 |
+
"step": 1300
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.08384,
|
| 936 |
+
"grad_norm": 1.0796362161636353,
|
| 937 |
+
"learning_rate": 5.585662470663538e-06,
|
| 938 |
+
"loss": 0.6652,
|
| 939 |
+
"step": 1310
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.08448,
|
| 943 |
+
"grad_norm": 2.0007293224334717,
|
| 944 |
+
"learning_rate": 5.628333688926819e-06,
|
| 945 |
+
"loss": 0.6573,
|
| 946 |
+
"step": 1320
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.08512,
|
| 950 |
+
"grad_norm": 1.3140546083450317,
|
| 951 |
+
"learning_rate": 5.671004907190101e-06,
|
| 952 |
+
"loss": 0.6642,
|
| 953 |
+
"step": 1330
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.08576,
|
| 957 |
+
"grad_norm": 1.6801965236663818,
|
| 958 |
+
"learning_rate": 5.7136761254533826e-06,
|
| 959 |
+
"loss": 0.6527,
|
| 960 |
+
"step": 1340
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.0864,
|
| 964 |
+
"grad_norm": 1.9289913177490234,
|
| 965 |
+
"learning_rate": 5.756347343716664e-06,
|
| 966 |
+
"loss": 0.6567,
|
| 967 |
+
"step": 1350
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 0.08704,
|
| 971 |
+
"grad_norm": 1.420436143875122,
|
| 972 |
+
"learning_rate": 5.799018561979946e-06,
|
| 973 |
+
"loss": 0.6479,
|
| 974 |
+
"step": 1360
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.08768,
|
| 978 |
+
"grad_norm": 1.3068914413452148,
|
| 979 |
+
"learning_rate": 5.841689780243226e-06,
|
| 980 |
+
"loss": 0.6522,
|
| 981 |
+
"step": 1370
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 0.08832,
|
| 985 |
+
"grad_norm": 1.8793392181396484,
|
| 986 |
+
"learning_rate": 5.884360998506507e-06,
|
| 987 |
+
"loss": 0.6491,
|
| 988 |
+
"step": 1380
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 0.08896,
|
| 992 |
+
"grad_norm": 1.3998247385025024,
|
| 993 |
+
"learning_rate": 5.927032216769789e-06,
|
| 994 |
+
"loss": 0.6638,
|
| 995 |
+
"step": 1390
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 0.0896,
|
| 999 |
+
"grad_norm": 1.371824026107788,
|
| 1000 |
+
"learning_rate": 5.96970343503307e-06,
|
| 1001 |
+
"loss": 0.6517,
|
| 1002 |
+
"step": 1400
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.09024,
|
| 1006 |
+
"grad_norm": 2.0050251483917236,
|
| 1007 |
+
"learning_rate": 6.012374653296352e-06,
|
| 1008 |
+
"loss": 0.6556,
|
| 1009 |
+
"step": 1410
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.09088,
|
| 1013 |
+
"grad_norm": 1.499855399131775,
|
| 1014 |
+
"learning_rate": 6.0550458715596335e-06,
|
| 1015 |
+
"loss": 0.6597,
|
| 1016 |
+
"step": 1420
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.09152,
|
| 1020 |
+
"grad_norm": 1.2066516876220703,
|
| 1021 |
+
"learning_rate": 6.097717089822915e-06,
|
| 1022 |
+
"loss": 0.6546,
|
| 1023 |
+
"step": 1430
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.09216,
|
| 1027 |
+
"grad_norm": 1.3586395978927612,
|
| 1028 |
+
"learning_rate": 6.140388308086196e-06,
|
| 1029 |
+
"loss": 0.6434,
|
| 1030 |
+
"step": 1440
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.0928,
|
| 1034 |
+
"grad_norm": 1.1413036584854126,
|
| 1035 |
+
"learning_rate": 6.183059526349477e-06,
|
| 1036 |
+
"loss": 0.6554,
|
| 1037 |
+
"step": 1450
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.09344,
|
| 1041 |
+
"grad_norm": 1.1844472885131836,
|
| 1042 |
+
"learning_rate": 6.225730744612759e-06,
|
| 1043 |
+
"loss": 0.6357,
|
| 1044 |
+
"step": 1460
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.09408,
|
| 1048 |
+
"grad_norm": 1.7203835248947144,
|
| 1049 |
+
"learning_rate": 6.2684019628760406e-06,
|
| 1050 |
+
"loss": 0.6546,
|
| 1051 |
+
"step": 1470
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.09472,
|
| 1055 |
+
"grad_norm": 1.1869184970855713,
|
| 1056 |
+
"learning_rate": 6.311073181139322e-06,
|
| 1057 |
+
"loss": 0.6376,
|
| 1058 |
+
"step": 1480
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.09536,
|
| 1062 |
+
"grad_norm": 1.3234835863113403,
|
| 1063 |
+
"learning_rate": 6.353744399402604e-06,
|
| 1064 |
+
"loss": 0.6494,
|
| 1065 |
+
"step": 1490
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.096,
|
| 1069 |
+
"grad_norm": 1.1924010515213013,
|
| 1070 |
+
"learning_rate": 6.396415617665885e-06,
|
| 1071 |
+
"loss": 0.652,
|
| 1072 |
+
"step": 1500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.09664,
|
| 1076 |
+
"grad_norm": 1.4505853652954102,
|
| 1077 |
+
"learning_rate": 6.439086835929166e-06,
|
| 1078 |
+
"loss": 0.6606,
|
| 1079 |
+
"step": 1510
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.09728,
|
| 1083 |
+
"grad_norm": 1.6351200342178345,
|
| 1084 |
+
"learning_rate": 6.481758054192448e-06,
|
| 1085 |
+
"loss": 0.6567,
|
| 1086 |
+
"step": 1520
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.09792,
|
| 1090 |
+
"grad_norm": 1.4423909187316895,
|
| 1091 |
+
"learning_rate": 6.524429272455729e-06,
|
| 1092 |
+
"loss": 0.653,
|
| 1093 |
+
"step": 1530
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 0.09856,
|
| 1097 |
+
"grad_norm": 0.8749169111251831,
|
| 1098 |
+
"learning_rate": 6.567100490719011e-06,
|
| 1099 |
+
"loss": 0.6417,
|
| 1100 |
+
"step": 1540
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 0.0992,
|
| 1104 |
+
"grad_norm": 1.2707208395004272,
|
| 1105 |
+
"learning_rate": 6.609771708982292e-06,
|
| 1106 |
+
"loss": 0.6563,
|
| 1107 |
+
"step": 1550
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 0.09984,
|
| 1111 |
+
"grad_norm": 1.163254976272583,
|
| 1112 |
+
"learning_rate": 6.652442927245574e-06,
|
| 1113 |
+
"loss": 0.6425,
|
| 1114 |
+
"step": 1560
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 0.10048,
|
| 1118 |
+
"grad_norm": 1.4674128293991089,
|
| 1119 |
+
"learning_rate": 6.6951141455088555e-06,
|
| 1120 |
+
"loss": 0.6561,
|
| 1121 |
+
"step": 1570
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 0.10112,
|
| 1125 |
+
"grad_norm": 1.6261919736862183,
|
| 1126 |
+
"learning_rate": 6.737785363772135e-06,
|
| 1127 |
+
"loss": 0.6473,
|
| 1128 |
+
"step": 1580
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 0.10176,
|
| 1132 |
+
"grad_norm": 1.2146090269088745,
|
| 1133 |
+
"learning_rate": 6.780456582035417e-06,
|
| 1134 |
+
"loss": 0.6467,
|
| 1135 |
+
"step": 1590
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 0.1024,
|
| 1139 |
+
"grad_norm": 1.3596712350845337,
|
| 1140 |
+
"learning_rate": 6.8231278002986985e-06,
|
| 1141 |
+
"loss": 0.6431,
|
| 1142 |
+
"step": 1600
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 0.10304,
|
| 1146 |
+
"grad_norm": 1.7861741781234741,
|
| 1147 |
+
"learning_rate": 6.86579901856198e-06,
|
| 1148 |
+
"loss": 0.6586,
|
| 1149 |
+
"step": 1610
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.10368,
|
| 1153 |
+
"grad_norm": 1.456587553024292,
|
| 1154 |
+
"learning_rate": 6.908470236825262e-06,
|
| 1155 |
+
"loss": 0.6496,
|
| 1156 |
+
"step": 1620
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 0.10432,
|
| 1160 |
+
"grad_norm": 1.5652499198913574,
|
| 1161 |
+
"learning_rate": 6.951141455088543e-06,
|
| 1162 |
+
"loss": 0.6436,
|
| 1163 |
+
"step": 1630
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 0.10496,
|
| 1167 |
+
"grad_norm": 1.1411017179489136,
|
| 1168 |
+
"learning_rate": 6.993812673351825e-06,
|
| 1169 |
+
"loss": 0.6572,
|
| 1170 |
+
"step": 1640
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 0.1056,
|
| 1174 |
+
"grad_norm": 1.3867056369781494,
|
| 1175 |
+
"learning_rate": 7.036483891615106e-06,
|
| 1176 |
+
"loss": 0.6425,
|
| 1177 |
+
"step": 1650
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 0.10624,
|
| 1181 |
+
"grad_norm": 1.2714463472366333,
|
| 1182 |
+
"learning_rate": 7.079155109878387e-06,
|
| 1183 |
+
"loss": 0.6558,
|
| 1184 |
+
"step": 1660
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 0.10688,
|
| 1188 |
+
"grad_norm": 1.32753586769104,
|
| 1189 |
+
"learning_rate": 7.121826328141669e-06,
|
| 1190 |
+
"loss": 0.6537,
|
| 1191 |
+
"step": 1670
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 0.10752,
|
| 1195 |
+
"grad_norm": 1.5559014081954956,
|
| 1196 |
+
"learning_rate": 7.16449754640495e-06,
|
| 1197 |
+
"loss": 0.6613,
|
| 1198 |
+
"step": 1680
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 0.10816,
|
| 1202 |
+
"grad_norm": 1.1726677417755127,
|
| 1203 |
+
"learning_rate": 7.207168764668232e-06,
|
| 1204 |
+
"loss": 0.6516,
|
| 1205 |
+
"step": 1690
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 0.1088,
|
| 1209 |
+
"grad_norm": 1.3453316688537598,
|
| 1210 |
+
"learning_rate": 7.2498399829315135e-06,
|
| 1211 |
+
"loss": 0.6455,
|
| 1212 |
+
"step": 1700
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 0.10944,
|
| 1216 |
+
"grad_norm": 1.6420834064483643,
|
| 1217 |
+
"learning_rate": 7.292511201194795e-06,
|
| 1218 |
+
"loss": 0.6457,
|
| 1219 |
+
"step": 1710
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 0.11008,
|
| 1223 |
+
"grad_norm": 1.0518358945846558,
|
| 1224 |
+
"learning_rate": 7.335182419458076e-06,
|
| 1225 |
+
"loss": 0.6315,
|
| 1226 |
+
"step": 1720
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 0.11072,
|
| 1230 |
+
"grad_norm": 1.4042829275131226,
|
| 1231 |
+
"learning_rate": 7.377853637721357e-06,
|
| 1232 |
+
"loss": 0.6445,
|
| 1233 |
+
"step": 1730
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 0.11136,
|
| 1237 |
+
"grad_norm": 1.2282599210739136,
|
| 1238 |
+
"learning_rate": 7.420524855984639e-06,
|
| 1239 |
+
"loss": 0.648,
|
| 1240 |
+
"step": 1740
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 0.112,
|
| 1244 |
+
"grad_norm": 1.2711114883422852,
|
| 1245 |
+
"learning_rate": 7.4631960742479205e-06,
|
| 1246 |
+
"loss": 0.6522,
|
| 1247 |
+
"step": 1750
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 0.11264,
|
| 1251 |
+
"grad_norm": 1.2256455421447754,
|
| 1252 |
+
"learning_rate": 7.505867292511202e-06,
|
| 1253 |
+
"loss": 0.6414,
|
| 1254 |
+
"step": 1760
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 0.11328,
|
| 1258 |
+
"grad_norm": 1.2845845222473145,
|
| 1259 |
+
"learning_rate": 7.548538510774484e-06,
|
| 1260 |
+
"loss": 0.6372,
|
| 1261 |
+
"step": 1770
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 0.11392,
|
| 1265 |
+
"grad_norm": 1.557308554649353,
|
| 1266 |
+
"learning_rate": 7.591209729037765e-06,
|
| 1267 |
+
"loss": 0.6472,
|
| 1268 |
+
"step": 1780
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 0.11456,
|
| 1272 |
+
"grad_norm": 1.5465153455734253,
|
| 1273 |
+
"learning_rate": 7.633880947301045e-06,
|
| 1274 |
+
"loss": 0.6387,
|
| 1275 |
+
"step": 1790
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 0.1152,
|
| 1279 |
+
"grad_norm": 1.0359770059585571,
|
| 1280 |
+
"learning_rate": 7.676552165564327e-06,
|
| 1281 |
+
"loss": 0.6335,
|
| 1282 |
+
"step": 1800
|
| 1283 |
+
}
|
| 1284 |
+
],
|
| 1285 |
+
"logging_steps": 10,
|
| 1286 |
+
"max_steps": 46875,
|
| 1287 |
+
"num_input_tokens_seen": 0,
|
| 1288 |
+
"num_train_epochs": 3,
|
| 1289 |
+
"save_steps": 200,
|
| 1290 |
+
"stateful_callbacks": {
|
| 1291 |
+
"EarlyStoppingCallback": {
|
| 1292 |
+
"args": {
|
| 1293 |
+
"early_stopping_patience": 3,
|
| 1294 |
+
"early_stopping_threshold": 0.0
|
| 1295 |
+
},
|
| 1296 |
+
"attributes": {
|
| 1297 |
+
"early_stopping_patience_counter": 0
|
| 1298 |
+
}
|
| 1299 |
+
},
|
| 1300 |
+
"TrainerControl": {
|
| 1301 |
+
"args": {
|
| 1302 |
+
"should_epoch_stop": false,
|
| 1303 |
+
"should_evaluate": false,
|
| 1304 |
+
"should_log": false,
|
| 1305 |
+
"should_save": true,
|
| 1306 |
+
"should_training_stop": false
|
| 1307 |
+
},
|
| 1308 |
+
"attributes": {}
|
| 1309 |
+
}
|
| 1310 |
+
},
|
| 1311 |
+
"total_flos": 1.514972078989632e+16,
|
| 1312 |
+
"train_batch_size": 32,
|
| 1313 |
+
"trial_name": null,
|
| 1314 |
+
"trial_params": null
|
| 1315 |
+
}
|
graphcodebert-rdrop/checkpoint-1800/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a0e6c50c4bb64305fe23a4e428084f9c072417915284e4ed838c0b852405eb80
|
| 3 |
+
size 5841
|
graphcodebert-rdrop/checkpoint-1800/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
graphcodebert-rdrop/checkpoint-2000/config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.2,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"gradient_checkpointing": false,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.2,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 514,
|
| 18 |
+
"model_type": "roberta",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"output_past": true,
|
| 22 |
+
"pad_token_id": 1,
|
| 23 |
+
"position_embedding_type": "absolute",
|
| 24 |
+
"transformers_version": "4.56.0",
|
| 25 |
+
"type_vocab_size": 1,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 50265
|
| 28 |
+
}
|
graphcodebert-rdrop/checkpoint-2000/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
graphcodebert-rdrop/checkpoint-2000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7db5571849e8e089f270ef2e4cf5cb4d9ef5cccb044fbe784913f0b02e58151b
|
| 3 |
+
size 498612824
|
graphcodebert-rdrop/checkpoint-2000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ada79dc0368fa04e1c8334181cba0de9346180550026b03291c9e953c46b8e45
|
| 3 |
+
size 4741923
|
graphcodebert-rdrop/checkpoint-2000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:06ecbf46522d24faf8472eb5b6502b67ca3d503299b0c81608339a5dd967a87a
|
| 3 |
+
size 14645
|
graphcodebert-rdrop/checkpoint-2000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4aa03f6e0cd07cf67ce1fbe3101d545f5771ef9148b9debf02b11cf6948da5c
|
| 3 |
+
size 1383
|
graphcodebert-rdrop/checkpoint-2000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:01081971fcef9864330db7d66449f3eacfc8bb4dd72c7f7888470931d0a4d9ae
|
| 3 |
+
size 1465
|
graphcodebert-rdrop/checkpoint-2000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
graphcodebert-rdrop/checkpoint-2000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
graphcodebert-rdrop/checkpoint-2000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 512,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 56 |
+
"trim_offsets": true,
|
| 57 |
+
"unk_token": "<unk>"
|
| 58 |
+
}
|
graphcodebert-rdrop/checkpoint-2000/trainer_state.json
ADDED
|
@@ -0,0 +1,1467 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 2000,
|
| 3 |
+
"best_metric": 0.797186028145479,
|
| 4 |
+
"best_model_checkpoint": "./output_checkpoints/graphcodebert-rdrop/checkpoint-2000",
|
| 5 |
+
"epoch": 0.128,
|
| 6 |
+
"eval_steps": 1000,
|
| 7 |
+
"global_step": 2000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00064,
|
| 14 |
+
"grad_norm": 1.8688431978225708,
|
| 15 |
+
"learning_rate": 3.840409643695328e-08,
|
| 16 |
+
"loss": 0.7168,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.00128,
|
| 21 |
+
"grad_norm": 1.835353970527649,
|
| 22 |
+
"learning_rate": 8.10753147002347e-08,
|
| 23 |
+
"loss": 0.7179,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00192,
|
| 28 |
+
"grad_norm": 1.2541388273239136,
|
| 29 |
+
"learning_rate": 1.2374653296351612e-07,
|
| 30 |
+
"loss": 0.7177,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.00256,
|
| 35 |
+
"grad_norm": 1.1381633281707764,
|
| 36 |
+
"learning_rate": 1.6641775122679754e-07,
|
| 37 |
+
"loss": 0.7312,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0032,
|
| 42 |
+
"grad_norm": 1.6313824653625488,
|
| 43 |
+
"learning_rate": 2.0908896949007894e-07,
|
| 44 |
+
"loss": 0.7167,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.00384,
|
| 49 |
+
"grad_norm": 1.092464566230774,
|
| 50 |
+
"learning_rate": 2.517601877533604e-07,
|
| 51 |
+
"loss": 0.7201,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.00448,
|
| 56 |
+
"grad_norm": 1.164099097251892,
|
| 57 |
+
"learning_rate": 2.944314060166418e-07,
|
| 58 |
+
"loss": 0.7195,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.00512,
|
| 63 |
+
"grad_norm": 1.9283920526504517,
|
| 64 |
+
"learning_rate": 3.371026242799232e-07,
|
| 65 |
+
"loss": 0.7223,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.00576,
|
| 70 |
+
"grad_norm": 1.8716129064559937,
|
| 71 |
+
"learning_rate": 3.7977384254320464e-07,
|
| 72 |
+
"loss": 0.7151,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0064,
|
| 77 |
+
"grad_norm": 1.5499262809753418,
|
| 78 |
+
"learning_rate": 4.22445060806486e-07,
|
| 79 |
+
"loss": 0.7211,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.00704,
|
| 84 |
+
"grad_norm": 1.375114917755127,
|
| 85 |
+
"learning_rate": 4.651162790697675e-07,
|
| 86 |
+
"loss": 0.7247,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.00768,
|
| 91 |
+
"grad_norm": 1.384915828704834,
|
| 92 |
+
"learning_rate": 5.077874973330489e-07,
|
| 93 |
+
"loss": 0.7156,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.00832,
|
| 98 |
+
"grad_norm": 1.6376659870147705,
|
| 99 |
+
"learning_rate": 5.504587155963304e-07,
|
| 100 |
+
"loss": 0.7196,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.00896,
|
| 105 |
+
"grad_norm": 1.709489345550537,
|
| 106 |
+
"learning_rate": 5.931299338596117e-07,
|
| 107 |
+
"loss": 0.7215,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0096,
|
| 112 |
+
"grad_norm": 1.350512146949768,
|
| 113 |
+
"learning_rate": 6.358011521228932e-07,
|
| 114 |
+
"loss": 0.71,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.01024,
|
| 119 |
+
"grad_norm": 2.33050537109375,
|
| 120 |
+
"learning_rate": 6.784723703861745e-07,
|
| 121 |
+
"loss": 0.7191,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.01088,
|
| 126 |
+
"grad_norm": 1.0042874813079834,
|
| 127 |
+
"learning_rate": 7.21143588649456e-07,
|
| 128 |
+
"loss": 0.72,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.01152,
|
| 133 |
+
"grad_norm": 1.1835744380950928,
|
| 134 |
+
"learning_rate": 7.638148069127374e-07,
|
| 135 |
+
"loss": 0.7122,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.01216,
|
| 140 |
+
"grad_norm": 1.949506402015686,
|
| 141 |
+
"learning_rate": 8.064860251760189e-07,
|
| 142 |
+
"loss": 0.7091,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.0128,
|
| 147 |
+
"grad_norm": 1.139626383781433,
|
| 148 |
+
"learning_rate": 8.491572434393003e-07,
|
| 149 |
+
"loss": 0.7046,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.01344,
|
| 154 |
+
"grad_norm": 1.7734779119491577,
|
| 155 |
+
"learning_rate": 8.918284617025817e-07,
|
| 156 |
+
"loss": 0.7135,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.01408,
|
| 161 |
+
"grad_norm": 1.3442974090576172,
|
| 162 |
+
"learning_rate": 9.344996799658632e-07,
|
| 163 |
+
"loss": 0.7127,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.01472,
|
| 168 |
+
"grad_norm": 1.6148335933685303,
|
| 169 |
+
"learning_rate": 9.771708982291445e-07,
|
| 170 |
+
"loss": 0.7168,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.01536,
|
| 175 |
+
"grad_norm": 0.9220213294029236,
|
| 176 |
+
"learning_rate": 1.0198421164924258e-06,
|
| 177 |
+
"loss": 0.7109,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.016,
|
| 182 |
+
"grad_norm": 1.5631815195083618,
|
| 183 |
+
"learning_rate": 1.0625133347557074e-06,
|
| 184 |
+
"loss": 0.7149,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.01664,
|
| 189 |
+
"grad_norm": 1.5052822828292847,
|
| 190 |
+
"learning_rate": 1.1051845530189888e-06,
|
| 191 |
+
"loss": 0.7112,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.01728,
|
| 196 |
+
"grad_norm": 1.6714903116226196,
|
| 197 |
+
"learning_rate": 1.1478557712822702e-06,
|
| 198 |
+
"loss": 0.7128,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.01792,
|
| 203 |
+
"grad_norm": 1.6667262315750122,
|
| 204 |
+
"learning_rate": 1.1905269895455517e-06,
|
| 205 |
+
"loss": 0.7091,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.01856,
|
| 210 |
+
"grad_norm": 1.6861907243728638,
|
| 211 |
+
"learning_rate": 1.233198207808833e-06,
|
| 212 |
+
"loss": 0.7078,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0192,
|
| 217 |
+
"grad_norm": 1.6259212493896484,
|
| 218 |
+
"learning_rate": 1.2758694260721145e-06,
|
| 219 |
+
"loss": 0.7087,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.01984,
|
| 224 |
+
"grad_norm": 1.207320213317871,
|
| 225 |
+
"learning_rate": 1.318540644335396e-06,
|
| 226 |
+
"loss": 0.7087,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.02048,
|
| 231 |
+
"grad_norm": 1.087124228477478,
|
| 232 |
+
"learning_rate": 1.3612118625986772e-06,
|
| 233 |
+
"loss": 0.7174,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.02112,
|
| 238 |
+
"grad_norm": 1.5668721199035645,
|
| 239 |
+
"learning_rate": 1.4038830808619588e-06,
|
| 240 |
+
"loss": 0.7066,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.02176,
|
| 245 |
+
"grad_norm": 1.5332859754562378,
|
| 246 |
+
"learning_rate": 1.4465542991252401e-06,
|
| 247 |
+
"loss": 0.7115,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0224,
|
| 252 |
+
"grad_norm": 2.092994451522827,
|
| 253 |
+
"learning_rate": 1.4892255173885215e-06,
|
| 254 |
+
"loss": 0.7085,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.02304,
|
| 259 |
+
"grad_norm": 1.266851544380188,
|
| 260 |
+
"learning_rate": 1.531896735651803e-06,
|
| 261 |
+
"loss": 0.7091,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.02368,
|
| 266 |
+
"grad_norm": 1.290616512298584,
|
| 267 |
+
"learning_rate": 1.5745679539150842e-06,
|
| 268 |
+
"loss": 0.7103,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.02432,
|
| 273 |
+
"grad_norm": 1.2045000791549683,
|
| 274 |
+
"learning_rate": 1.6172391721783658e-06,
|
| 275 |
+
"loss": 0.705,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.02496,
|
| 280 |
+
"grad_norm": 1.828832983970642,
|
| 281 |
+
"learning_rate": 1.6599103904416472e-06,
|
| 282 |
+
"loss": 0.7038,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.0256,
|
| 287 |
+
"grad_norm": 1.7793196439743042,
|
| 288 |
+
"learning_rate": 1.7025816087049288e-06,
|
| 289 |
+
"loss": 0.7085,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.02624,
|
| 294 |
+
"grad_norm": 0.8949472904205322,
|
| 295 |
+
"learning_rate": 1.7452528269682101e-06,
|
| 296 |
+
"loss": 0.7045,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.02688,
|
| 301 |
+
"grad_norm": 1.3071945905685425,
|
| 302 |
+
"learning_rate": 1.7879240452314913e-06,
|
| 303 |
+
"loss": 0.702,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.02752,
|
| 308 |
+
"grad_norm": 1.7497148513793945,
|
| 309 |
+
"learning_rate": 1.8305952634947729e-06,
|
| 310 |
+
"loss": 0.7073,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.02816,
|
| 315 |
+
"grad_norm": 1.0856297016143799,
|
| 316 |
+
"learning_rate": 1.8732664817580542e-06,
|
| 317 |
+
"loss": 0.7008,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.0288,
|
| 322 |
+
"grad_norm": 1.369019627571106,
|
| 323 |
+
"learning_rate": 1.915937700021336e-06,
|
| 324 |
+
"loss": 0.7059,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.02944,
|
| 329 |
+
"grad_norm": 2.0961010456085205,
|
| 330 |
+
"learning_rate": 1.958608918284617e-06,
|
| 331 |
+
"loss": 0.6952,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.03008,
|
| 336 |
+
"grad_norm": 1.6076347827911377,
|
| 337 |
+
"learning_rate": 2.0012801365478988e-06,
|
| 338 |
+
"loss": 0.7074,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.03072,
|
| 343 |
+
"grad_norm": 1.3129311800003052,
|
| 344 |
+
"learning_rate": 2.04395135481118e-06,
|
| 345 |
+
"loss": 0.6994,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.03136,
|
| 350 |
+
"grad_norm": 1.3621476888656616,
|
| 351 |
+
"learning_rate": 2.0866225730744615e-06,
|
| 352 |
+
"loss": 0.696,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.032,
|
| 357 |
+
"grad_norm": 0.9786806106567383,
|
| 358 |
+
"learning_rate": 2.129293791337743e-06,
|
| 359 |
+
"loss": 0.7011,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.03264,
|
| 364 |
+
"grad_norm": 1.4611176252365112,
|
| 365 |
+
"learning_rate": 2.1719650096010242e-06,
|
| 366 |
+
"loss": 0.699,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.03328,
|
| 371 |
+
"grad_norm": 1.0675945281982422,
|
| 372 |
+
"learning_rate": 2.214636227864306e-06,
|
| 373 |
+
"loss": 0.6906,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.03392,
|
| 378 |
+
"grad_norm": 1.6564017534255981,
|
| 379 |
+
"learning_rate": 2.257307446127587e-06,
|
| 380 |
+
"loss": 0.6923,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.03456,
|
| 385 |
+
"grad_norm": 1.228119134902954,
|
| 386 |
+
"learning_rate": 2.2999786643908685e-06,
|
| 387 |
+
"loss": 0.6928,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.0352,
|
| 392 |
+
"grad_norm": 1.9868593215942383,
|
| 393 |
+
"learning_rate": 2.34264988265415e-06,
|
| 394 |
+
"loss": 0.6912,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.03584,
|
| 399 |
+
"grad_norm": 1.2531176805496216,
|
| 400 |
+
"learning_rate": 2.3853211009174317e-06,
|
| 401 |
+
"loss": 0.6923,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.03648,
|
| 406 |
+
"grad_norm": 1.413602352142334,
|
| 407 |
+
"learning_rate": 2.427992319180713e-06,
|
| 408 |
+
"loss": 0.6993,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.03712,
|
| 413 |
+
"grad_norm": 0.9655390977859497,
|
| 414 |
+
"learning_rate": 2.470663537443994e-06,
|
| 415 |
+
"loss": 0.7017,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.03776,
|
| 420 |
+
"grad_norm": 1.962438941001892,
|
| 421 |
+
"learning_rate": 2.5133347557072756e-06,
|
| 422 |
+
"loss": 0.6933,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.0384,
|
| 427 |
+
"grad_norm": 1.1099931001663208,
|
| 428 |
+
"learning_rate": 2.556005973970557e-06,
|
| 429 |
+
"loss": 0.6925,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.03904,
|
| 434 |
+
"grad_norm": 1.1766624450683594,
|
| 435 |
+
"learning_rate": 2.5986771922338383e-06,
|
| 436 |
+
"loss": 0.6894,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.03968,
|
| 441 |
+
"grad_norm": 1.574353814125061,
|
| 442 |
+
"learning_rate": 2.64134841049712e-06,
|
| 443 |
+
"loss": 0.6923,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.04032,
|
| 448 |
+
"grad_norm": 1.194074273109436,
|
| 449 |
+
"learning_rate": 2.6840196287604015e-06,
|
| 450 |
+
"loss": 0.6855,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.04096,
|
| 455 |
+
"grad_norm": 1.1750593185424805,
|
| 456 |
+
"learning_rate": 2.7266908470236826e-06,
|
| 457 |
+
"loss": 0.7002,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.0416,
|
| 462 |
+
"grad_norm": 1.3165347576141357,
|
| 463 |
+
"learning_rate": 2.7693620652869642e-06,
|
| 464 |
+
"loss": 0.6893,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.04224,
|
| 469 |
+
"grad_norm": 1.1079384088516235,
|
| 470 |
+
"learning_rate": 2.812033283550246e-06,
|
| 471 |
+
"loss": 0.6852,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.04288,
|
| 476 |
+
"grad_norm": 1.231327772140503,
|
| 477 |
+
"learning_rate": 2.8547045018135274e-06,
|
| 478 |
+
"loss": 0.695,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.04352,
|
| 483 |
+
"grad_norm": 1.966036319732666,
|
| 484 |
+
"learning_rate": 2.897375720076808e-06,
|
| 485 |
+
"loss": 0.6833,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.04416,
|
| 490 |
+
"grad_norm": 1.0460131168365479,
|
| 491 |
+
"learning_rate": 2.9400469383400897e-06,
|
| 492 |
+
"loss": 0.6802,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.0448,
|
| 497 |
+
"grad_norm": 1.2445200681686401,
|
| 498 |
+
"learning_rate": 2.9827181566033713e-06,
|
| 499 |
+
"loss": 0.6883,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.04544,
|
| 504 |
+
"grad_norm": 1.1759636402130127,
|
| 505 |
+
"learning_rate": 3.025389374866653e-06,
|
| 506 |
+
"loss": 0.682,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.04608,
|
| 511 |
+
"grad_norm": 1.2478561401367188,
|
| 512 |
+
"learning_rate": 3.068060593129934e-06,
|
| 513 |
+
"loss": 0.6802,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.04672,
|
| 518 |
+
"grad_norm": 0.8945108652114868,
|
| 519 |
+
"learning_rate": 3.1107318113932156e-06,
|
| 520 |
+
"loss": 0.6873,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.04736,
|
| 525 |
+
"grad_norm": 1.6810317039489746,
|
| 526 |
+
"learning_rate": 3.153403029656497e-06,
|
| 527 |
+
"loss": 0.6828,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.048,
|
| 532 |
+
"grad_norm": 2.882283926010132,
|
| 533 |
+
"learning_rate": 3.1960742479197783e-06,
|
| 534 |
+
"loss": 0.6797,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.04864,
|
| 539 |
+
"grad_norm": 1.575766921043396,
|
| 540 |
+
"learning_rate": 3.23874546618306e-06,
|
| 541 |
+
"loss": 0.6826,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.04928,
|
| 546 |
+
"grad_norm": 1.8044737577438354,
|
| 547 |
+
"learning_rate": 3.2814166844463415e-06,
|
| 548 |
+
"loss": 0.6837,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.04992,
|
| 553 |
+
"grad_norm": 1.4755513668060303,
|
| 554 |
+
"learning_rate": 3.324087902709623e-06,
|
| 555 |
+
"loss": 0.6815,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.05056,
|
| 560 |
+
"grad_norm": 2.0735654830932617,
|
| 561 |
+
"learning_rate": 3.3667591209729038e-06,
|
| 562 |
+
"loss": 0.6773,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.0512,
|
| 567 |
+
"grad_norm": 0.9823655486106873,
|
| 568 |
+
"learning_rate": 3.4094303392361854e-06,
|
| 569 |
+
"loss": 0.6689,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.05184,
|
| 574 |
+
"grad_norm": 1.0731920003890991,
|
| 575 |
+
"learning_rate": 3.452101557499467e-06,
|
| 576 |
+
"loss": 0.6689,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.05248,
|
| 581 |
+
"grad_norm": 1.1081087589263916,
|
| 582 |
+
"learning_rate": 3.494772775762748e-06,
|
| 583 |
+
"loss": 0.6694,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.05312,
|
| 588 |
+
"grad_norm": 1.2130361795425415,
|
| 589 |
+
"learning_rate": 3.5374439940260297e-06,
|
| 590 |
+
"loss": 0.682,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.05376,
|
| 595 |
+
"grad_norm": 1.170345664024353,
|
| 596 |
+
"learning_rate": 3.5801152122893113e-06,
|
| 597 |
+
"loss": 0.6756,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.0544,
|
| 602 |
+
"grad_norm": 1.2610124349594116,
|
| 603 |
+
"learning_rate": 3.622786430552593e-06,
|
| 604 |
+
"loss": 0.67,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.05504,
|
| 609 |
+
"grad_norm": 1.0138545036315918,
|
| 610 |
+
"learning_rate": 3.665457648815874e-06,
|
| 611 |
+
"loss": 0.6807,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.05568,
|
| 616 |
+
"grad_norm": 1.0937036275863647,
|
| 617 |
+
"learning_rate": 3.7081288670791556e-06,
|
| 618 |
+
"loss": 0.6774,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.05632,
|
| 623 |
+
"grad_norm": 1.119221568107605,
|
| 624 |
+
"learning_rate": 3.750800085342437e-06,
|
| 625 |
+
"loss": 0.6728,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.05696,
|
| 630 |
+
"grad_norm": 1.9351646900177002,
|
| 631 |
+
"learning_rate": 3.793471303605718e-06,
|
| 632 |
+
"loss": 0.6664,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.0576,
|
| 637 |
+
"grad_norm": 1.6208665370941162,
|
| 638 |
+
"learning_rate": 3.836142521869e-06,
|
| 639 |
+
"loss": 0.6653,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.05824,
|
| 644 |
+
"grad_norm": 1.3460793495178223,
|
| 645 |
+
"learning_rate": 3.8788137401322815e-06,
|
| 646 |
+
"loss": 0.6651,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.05888,
|
| 651 |
+
"grad_norm": 1.8987629413604736,
|
| 652 |
+
"learning_rate": 3.921484958395563e-06,
|
| 653 |
+
"loss": 0.6694,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.05952,
|
| 658 |
+
"grad_norm": 1.3637055158615112,
|
| 659 |
+
"learning_rate": 3.964156176658844e-06,
|
| 660 |
+
"loss": 0.6691,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.06016,
|
| 665 |
+
"grad_norm": 1.3319580554962158,
|
| 666 |
+
"learning_rate": 4.006827394922125e-06,
|
| 667 |
+
"loss": 0.6749,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.0608,
|
| 672 |
+
"grad_norm": 1.3574756383895874,
|
| 673 |
+
"learning_rate": 4.049498613185407e-06,
|
| 674 |
+
"loss": 0.6688,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.06144,
|
| 679 |
+
"grad_norm": 1.136319637298584,
|
| 680 |
+
"learning_rate": 4.092169831448688e-06,
|
| 681 |
+
"loss": 0.6657,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.06208,
|
| 686 |
+
"grad_norm": 1.7666376829147339,
|
| 687 |
+
"learning_rate": 4.134841049711969e-06,
|
| 688 |
+
"loss": 0.6667,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.06272,
|
| 693 |
+
"grad_norm": 1.332014560699463,
|
| 694 |
+
"learning_rate": 4.177512267975251e-06,
|
| 695 |
+
"loss": 0.6715,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.06336,
|
| 700 |
+
"grad_norm": 1.3127824068069458,
|
| 701 |
+
"learning_rate": 4.220183486238532e-06,
|
| 702 |
+
"loss": 0.6737,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.064,
|
| 707 |
+
"grad_norm": 1.293678641319275,
|
| 708 |
+
"learning_rate": 4.262854704501814e-06,
|
| 709 |
+
"loss": 0.6557,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.064,
|
| 714 |
+
"eval_accuracy": 0.76589,
|
| 715 |
+
"eval_loss": 0.6282070279121399,
|
| 716 |
+
"eval_macro_f1": 0.7657238614598489,
|
| 717 |
+
"eval_precision": 0.7708542319194018,
|
| 718 |
+
"eval_recall": 0.7687514011652705,
|
| 719 |
+
"eval_runtime": 806.5969,
|
| 720 |
+
"eval_samples_per_second": 123.978,
|
| 721 |
+
"eval_steps_per_second": 1.938,
|
| 722 |
+
"step": 1000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.06464,
|
| 726 |
+
"grad_norm": 1.2433074712753296,
|
| 727 |
+
"learning_rate": 4.3055259227650956e-06,
|
| 728 |
+
"loss": 0.6583,
|
| 729 |
+
"step": 1010
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.06528,
|
| 733 |
+
"grad_norm": 1.5115686655044556,
|
| 734 |
+
"learning_rate": 4.348197141028377e-06,
|
| 735 |
+
"loss": 0.6659,
|
| 736 |
+
"step": 1020
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.06592,
|
| 740 |
+
"grad_norm": 1.3560088872909546,
|
| 741 |
+
"learning_rate": 4.390868359291658e-06,
|
| 742 |
+
"loss": 0.6675,
|
| 743 |
+
"step": 1030
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.06656,
|
| 747 |
+
"grad_norm": 0.8229928016662598,
|
| 748 |
+
"learning_rate": 4.4335395775549394e-06,
|
| 749 |
+
"loss": 0.6629,
|
| 750 |
+
"step": 1040
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.0672,
|
| 754 |
+
"grad_norm": 1.5138262510299683,
|
| 755 |
+
"learning_rate": 4.476210795818221e-06,
|
| 756 |
+
"loss": 0.6635,
|
| 757 |
+
"step": 1050
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 0.06784,
|
| 761 |
+
"grad_norm": 1.1732233762741089,
|
| 762 |
+
"learning_rate": 4.518882014081503e-06,
|
| 763 |
+
"loss": 0.6531,
|
| 764 |
+
"step": 1060
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.06848,
|
| 768 |
+
"grad_norm": 1.6118066310882568,
|
| 769 |
+
"learning_rate": 4.561553232344783e-06,
|
| 770 |
+
"loss": 0.6644,
|
| 771 |
+
"step": 1070
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 0.06912,
|
| 775 |
+
"grad_norm": 0.9575800895690918,
|
| 776 |
+
"learning_rate": 4.604224450608065e-06,
|
| 777 |
+
"loss": 0.6542,
|
| 778 |
+
"step": 1080
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.06976,
|
| 782 |
+
"grad_norm": 1.157271146774292,
|
| 783 |
+
"learning_rate": 4.6468956688713465e-06,
|
| 784 |
+
"loss": 0.6505,
|
| 785 |
+
"step": 1090
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.0704,
|
| 789 |
+
"grad_norm": 1.4502589702606201,
|
| 790 |
+
"learning_rate": 4.689566887134628e-06,
|
| 791 |
+
"loss": 0.6725,
|
| 792 |
+
"step": 1100
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 0.07104,
|
| 796 |
+
"grad_norm": 1.122251033782959,
|
| 797 |
+
"learning_rate": 4.73223810539791e-06,
|
| 798 |
+
"loss": 0.6625,
|
| 799 |
+
"step": 1110
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 0.07168,
|
| 803 |
+
"grad_norm": 0.9529566764831543,
|
| 804 |
+
"learning_rate": 4.774909323661191e-06,
|
| 805 |
+
"loss": 0.6684,
|
| 806 |
+
"step": 1120
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.07232,
|
| 810 |
+
"grad_norm": 1.6433424949645996,
|
| 811 |
+
"learning_rate": 4.817580541924473e-06,
|
| 812 |
+
"loss": 0.6599,
|
| 813 |
+
"step": 1130
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.07296,
|
| 817 |
+
"grad_norm": 1.6565576791763306,
|
| 818 |
+
"learning_rate": 4.8602517601877535e-06,
|
| 819 |
+
"loss": 0.6591,
|
| 820 |
+
"step": 1140
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 0.0736,
|
| 824 |
+
"grad_norm": 1.3955042362213135,
|
| 825 |
+
"learning_rate": 4.902922978451035e-06,
|
| 826 |
+
"loss": 0.6649,
|
| 827 |
+
"step": 1150
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 0.07424,
|
| 831 |
+
"grad_norm": 1.530819296836853,
|
| 832 |
+
"learning_rate": 4.945594196714317e-06,
|
| 833 |
+
"loss": 0.65,
|
| 834 |
+
"step": 1160
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.07488,
|
| 838 |
+
"grad_norm": 1.0729092359542847,
|
| 839 |
+
"learning_rate": 4.9882654149775974e-06,
|
| 840 |
+
"loss": 0.6603,
|
| 841 |
+
"step": 1170
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.07552,
|
| 845 |
+
"grad_norm": 1.6307648420333862,
|
| 846 |
+
"learning_rate": 5.030936633240879e-06,
|
| 847 |
+
"loss": 0.6702,
|
| 848 |
+
"step": 1180
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.07616,
|
| 852 |
+
"grad_norm": 1.324406623840332,
|
| 853 |
+
"learning_rate": 5.073607851504161e-06,
|
| 854 |
+
"loss": 0.6634,
|
| 855 |
+
"step": 1190
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 0.0768,
|
| 859 |
+
"grad_norm": 1.1919140815734863,
|
| 860 |
+
"learning_rate": 5.116279069767442e-06,
|
| 861 |
+
"loss": 0.6586,
|
| 862 |
+
"step": 1200
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 0.07744,
|
| 866 |
+
"grad_norm": 1.0658751726150513,
|
| 867 |
+
"learning_rate": 5.158950288030724e-06,
|
| 868 |
+
"loss": 0.6573,
|
| 869 |
+
"step": 1210
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 0.07808,
|
| 873 |
+
"grad_norm": 1.7762951850891113,
|
| 874 |
+
"learning_rate": 5.201621506294005e-06,
|
| 875 |
+
"loss": 0.6596,
|
| 876 |
+
"step": 1220
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 0.07872,
|
| 880 |
+
"grad_norm": 1.1911667585372925,
|
| 881 |
+
"learning_rate": 5.244292724557286e-06,
|
| 882 |
+
"loss": 0.651,
|
| 883 |
+
"step": 1230
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 0.07936,
|
| 887 |
+
"grad_norm": 1.442029595375061,
|
| 888 |
+
"learning_rate": 5.286963942820568e-06,
|
| 889 |
+
"loss": 0.6598,
|
| 890 |
+
"step": 1240
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.08,
|
| 894 |
+
"grad_norm": 0.9955100417137146,
|
| 895 |
+
"learning_rate": 5.329635161083849e-06,
|
| 896 |
+
"loss": 0.6587,
|
| 897 |
+
"step": 1250
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.08064,
|
| 901 |
+
"grad_norm": 1.1801563501358032,
|
| 902 |
+
"learning_rate": 5.372306379347131e-06,
|
| 903 |
+
"loss": 0.6669,
|
| 904 |
+
"step": 1260
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.08128,
|
| 908 |
+
"grad_norm": 1.8548506498336792,
|
| 909 |
+
"learning_rate": 5.414977597610412e-06,
|
| 910 |
+
"loss": 0.6534,
|
| 911 |
+
"step": 1270
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.08192,
|
| 915 |
+
"grad_norm": 1.8297194242477417,
|
| 916 |
+
"learning_rate": 5.457648815873694e-06,
|
| 917 |
+
"loss": 0.668,
|
| 918 |
+
"step": 1280
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 0.08256,
|
| 922 |
+
"grad_norm": 1.3471113443374634,
|
| 923 |
+
"learning_rate": 5.5003200341369755e-06,
|
| 924 |
+
"loss": 0.659,
|
| 925 |
+
"step": 1290
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 0.0832,
|
| 929 |
+
"grad_norm": 1.4015250205993652,
|
| 930 |
+
"learning_rate": 5.542991252400256e-06,
|
| 931 |
+
"loss": 0.6602,
|
| 932 |
+
"step": 1300
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.08384,
|
| 936 |
+
"grad_norm": 1.0796362161636353,
|
| 937 |
+
"learning_rate": 5.585662470663538e-06,
|
| 938 |
+
"loss": 0.6652,
|
| 939 |
+
"step": 1310
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.08448,
|
| 943 |
+
"grad_norm": 2.0007293224334717,
|
| 944 |
+
"learning_rate": 5.628333688926819e-06,
|
| 945 |
+
"loss": 0.6573,
|
| 946 |
+
"step": 1320
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.08512,
|
| 950 |
+
"grad_norm": 1.3140546083450317,
|
| 951 |
+
"learning_rate": 5.671004907190101e-06,
|
| 952 |
+
"loss": 0.6642,
|
| 953 |
+
"step": 1330
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.08576,
|
| 957 |
+
"grad_norm": 1.6801965236663818,
|
| 958 |
+
"learning_rate": 5.7136761254533826e-06,
|
| 959 |
+
"loss": 0.6527,
|
| 960 |
+
"step": 1340
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.0864,
|
| 964 |
+
"grad_norm": 1.9289913177490234,
|
| 965 |
+
"learning_rate": 5.756347343716664e-06,
|
| 966 |
+
"loss": 0.6567,
|
| 967 |
+
"step": 1350
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 0.08704,
|
| 971 |
+
"grad_norm": 1.420436143875122,
|
| 972 |
+
"learning_rate": 5.799018561979946e-06,
|
| 973 |
+
"loss": 0.6479,
|
| 974 |
+
"step": 1360
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.08768,
|
| 978 |
+
"grad_norm": 1.3068914413452148,
|
| 979 |
+
"learning_rate": 5.841689780243226e-06,
|
| 980 |
+
"loss": 0.6522,
|
| 981 |
+
"step": 1370
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 0.08832,
|
| 985 |
+
"grad_norm": 1.8793392181396484,
|
| 986 |
+
"learning_rate": 5.884360998506507e-06,
|
| 987 |
+
"loss": 0.6491,
|
| 988 |
+
"step": 1380
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 0.08896,
|
| 992 |
+
"grad_norm": 1.3998247385025024,
|
| 993 |
+
"learning_rate": 5.927032216769789e-06,
|
| 994 |
+
"loss": 0.6638,
|
| 995 |
+
"step": 1390
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 0.0896,
|
| 999 |
+
"grad_norm": 1.371824026107788,
|
| 1000 |
+
"learning_rate": 5.96970343503307e-06,
|
| 1001 |
+
"loss": 0.6517,
|
| 1002 |
+
"step": 1400
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.09024,
|
| 1006 |
+
"grad_norm": 2.0050251483917236,
|
| 1007 |
+
"learning_rate": 6.012374653296352e-06,
|
| 1008 |
+
"loss": 0.6556,
|
| 1009 |
+
"step": 1410
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.09088,
|
| 1013 |
+
"grad_norm": 1.499855399131775,
|
| 1014 |
+
"learning_rate": 6.0550458715596335e-06,
|
| 1015 |
+
"loss": 0.6597,
|
| 1016 |
+
"step": 1420
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.09152,
|
| 1020 |
+
"grad_norm": 1.2066516876220703,
|
| 1021 |
+
"learning_rate": 6.097717089822915e-06,
|
| 1022 |
+
"loss": 0.6546,
|
| 1023 |
+
"step": 1430
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.09216,
|
| 1027 |
+
"grad_norm": 1.3586395978927612,
|
| 1028 |
+
"learning_rate": 6.140388308086196e-06,
|
| 1029 |
+
"loss": 0.6434,
|
| 1030 |
+
"step": 1440
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.0928,
|
| 1034 |
+
"grad_norm": 1.1413036584854126,
|
| 1035 |
+
"learning_rate": 6.183059526349477e-06,
|
| 1036 |
+
"loss": 0.6554,
|
| 1037 |
+
"step": 1450
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.09344,
|
| 1041 |
+
"grad_norm": 1.1844472885131836,
|
| 1042 |
+
"learning_rate": 6.225730744612759e-06,
|
| 1043 |
+
"loss": 0.6357,
|
| 1044 |
+
"step": 1460
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.09408,
|
| 1048 |
+
"grad_norm": 1.7203835248947144,
|
| 1049 |
+
"learning_rate": 6.2684019628760406e-06,
|
| 1050 |
+
"loss": 0.6546,
|
| 1051 |
+
"step": 1470
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.09472,
|
| 1055 |
+
"grad_norm": 1.1869184970855713,
|
| 1056 |
+
"learning_rate": 6.311073181139322e-06,
|
| 1057 |
+
"loss": 0.6376,
|
| 1058 |
+
"step": 1480
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.09536,
|
| 1062 |
+
"grad_norm": 1.3234835863113403,
|
| 1063 |
+
"learning_rate": 6.353744399402604e-06,
|
| 1064 |
+
"loss": 0.6494,
|
| 1065 |
+
"step": 1490
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.096,
|
| 1069 |
+
"grad_norm": 1.1924010515213013,
|
| 1070 |
+
"learning_rate": 6.396415617665885e-06,
|
| 1071 |
+
"loss": 0.652,
|
| 1072 |
+
"step": 1500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.09664,
|
| 1076 |
+
"grad_norm": 1.4505853652954102,
|
| 1077 |
+
"learning_rate": 6.439086835929166e-06,
|
| 1078 |
+
"loss": 0.6606,
|
| 1079 |
+
"step": 1510
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.09728,
|
| 1083 |
+
"grad_norm": 1.6351200342178345,
|
| 1084 |
+
"learning_rate": 6.481758054192448e-06,
|
| 1085 |
+
"loss": 0.6567,
|
| 1086 |
+
"step": 1520
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.09792,
|
| 1090 |
+
"grad_norm": 1.4423909187316895,
|
| 1091 |
+
"learning_rate": 6.524429272455729e-06,
|
| 1092 |
+
"loss": 0.653,
|
| 1093 |
+
"step": 1530
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 0.09856,
|
| 1097 |
+
"grad_norm": 0.8749169111251831,
|
| 1098 |
+
"learning_rate": 6.567100490719011e-06,
|
| 1099 |
+
"loss": 0.6417,
|
| 1100 |
+
"step": 1540
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 0.0992,
|
| 1104 |
+
"grad_norm": 1.2707208395004272,
|
| 1105 |
+
"learning_rate": 6.609771708982292e-06,
|
| 1106 |
+
"loss": 0.6563,
|
| 1107 |
+
"step": 1550
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 0.09984,
|
| 1111 |
+
"grad_norm": 1.163254976272583,
|
| 1112 |
+
"learning_rate": 6.652442927245574e-06,
|
| 1113 |
+
"loss": 0.6425,
|
| 1114 |
+
"step": 1560
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 0.10048,
|
| 1118 |
+
"grad_norm": 1.4674128293991089,
|
| 1119 |
+
"learning_rate": 6.6951141455088555e-06,
|
| 1120 |
+
"loss": 0.6561,
|
| 1121 |
+
"step": 1570
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 0.10112,
|
| 1125 |
+
"grad_norm": 1.6261919736862183,
|
| 1126 |
+
"learning_rate": 6.737785363772135e-06,
|
| 1127 |
+
"loss": 0.6473,
|
| 1128 |
+
"step": 1580
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 0.10176,
|
| 1132 |
+
"grad_norm": 1.2146090269088745,
|
| 1133 |
+
"learning_rate": 6.780456582035417e-06,
|
| 1134 |
+
"loss": 0.6467,
|
| 1135 |
+
"step": 1590
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 0.1024,
|
| 1139 |
+
"grad_norm": 1.3596712350845337,
|
| 1140 |
+
"learning_rate": 6.8231278002986985e-06,
|
| 1141 |
+
"loss": 0.6431,
|
| 1142 |
+
"step": 1600
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 0.10304,
|
| 1146 |
+
"grad_norm": 1.7861741781234741,
|
| 1147 |
+
"learning_rate": 6.86579901856198e-06,
|
| 1148 |
+
"loss": 0.6586,
|
| 1149 |
+
"step": 1610
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.10368,
|
| 1153 |
+
"grad_norm": 1.456587553024292,
|
| 1154 |
+
"learning_rate": 6.908470236825262e-06,
|
| 1155 |
+
"loss": 0.6496,
|
| 1156 |
+
"step": 1620
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 0.10432,
|
| 1160 |
+
"grad_norm": 1.5652499198913574,
|
| 1161 |
+
"learning_rate": 6.951141455088543e-06,
|
| 1162 |
+
"loss": 0.6436,
|
| 1163 |
+
"step": 1630
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 0.10496,
|
| 1167 |
+
"grad_norm": 1.1411017179489136,
|
| 1168 |
+
"learning_rate": 6.993812673351825e-06,
|
| 1169 |
+
"loss": 0.6572,
|
| 1170 |
+
"step": 1640
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 0.1056,
|
| 1174 |
+
"grad_norm": 1.3867056369781494,
|
| 1175 |
+
"learning_rate": 7.036483891615106e-06,
|
| 1176 |
+
"loss": 0.6425,
|
| 1177 |
+
"step": 1650
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 0.10624,
|
| 1181 |
+
"grad_norm": 1.2714463472366333,
|
| 1182 |
+
"learning_rate": 7.079155109878387e-06,
|
| 1183 |
+
"loss": 0.6558,
|
| 1184 |
+
"step": 1660
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 0.10688,
|
| 1188 |
+
"grad_norm": 1.32753586769104,
|
| 1189 |
+
"learning_rate": 7.121826328141669e-06,
|
| 1190 |
+
"loss": 0.6537,
|
| 1191 |
+
"step": 1670
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 0.10752,
|
| 1195 |
+
"grad_norm": 1.5559014081954956,
|
| 1196 |
+
"learning_rate": 7.16449754640495e-06,
|
| 1197 |
+
"loss": 0.6613,
|
| 1198 |
+
"step": 1680
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 0.10816,
|
| 1202 |
+
"grad_norm": 1.1726677417755127,
|
| 1203 |
+
"learning_rate": 7.207168764668232e-06,
|
| 1204 |
+
"loss": 0.6516,
|
| 1205 |
+
"step": 1690
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 0.1088,
|
| 1209 |
+
"grad_norm": 1.3453316688537598,
|
| 1210 |
+
"learning_rate": 7.2498399829315135e-06,
|
| 1211 |
+
"loss": 0.6455,
|
| 1212 |
+
"step": 1700
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 0.10944,
|
| 1216 |
+
"grad_norm": 1.6420834064483643,
|
| 1217 |
+
"learning_rate": 7.292511201194795e-06,
|
| 1218 |
+
"loss": 0.6457,
|
| 1219 |
+
"step": 1710
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 0.11008,
|
| 1223 |
+
"grad_norm": 1.0518358945846558,
|
| 1224 |
+
"learning_rate": 7.335182419458076e-06,
|
| 1225 |
+
"loss": 0.6315,
|
| 1226 |
+
"step": 1720
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 0.11072,
|
| 1230 |
+
"grad_norm": 1.4042829275131226,
|
| 1231 |
+
"learning_rate": 7.377853637721357e-06,
|
| 1232 |
+
"loss": 0.6445,
|
| 1233 |
+
"step": 1730
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 0.11136,
|
| 1237 |
+
"grad_norm": 1.2282599210739136,
|
| 1238 |
+
"learning_rate": 7.420524855984639e-06,
|
| 1239 |
+
"loss": 0.648,
|
| 1240 |
+
"step": 1740
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 0.112,
|
| 1244 |
+
"grad_norm": 1.2711114883422852,
|
| 1245 |
+
"learning_rate": 7.4631960742479205e-06,
|
| 1246 |
+
"loss": 0.6522,
|
| 1247 |
+
"step": 1750
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 0.11264,
|
| 1251 |
+
"grad_norm": 1.2256455421447754,
|
| 1252 |
+
"learning_rate": 7.505867292511202e-06,
|
| 1253 |
+
"loss": 0.6414,
|
| 1254 |
+
"step": 1760
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 0.11328,
|
| 1258 |
+
"grad_norm": 1.2845845222473145,
|
| 1259 |
+
"learning_rate": 7.548538510774484e-06,
|
| 1260 |
+
"loss": 0.6372,
|
| 1261 |
+
"step": 1770
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 0.11392,
|
| 1265 |
+
"grad_norm": 1.557308554649353,
|
| 1266 |
+
"learning_rate": 7.591209729037765e-06,
|
| 1267 |
+
"loss": 0.6472,
|
| 1268 |
+
"step": 1780
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 0.11456,
|
| 1272 |
+
"grad_norm": 1.5465153455734253,
|
| 1273 |
+
"learning_rate": 7.633880947301045e-06,
|
| 1274 |
+
"loss": 0.6387,
|
| 1275 |
+
"step": 1790
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 0.1152,
|
| 1279 |
+
"grad_norm": 1.0359770059585571,
|
| 1280 |
+
"learning_rate": 7.676552165564327e-06,
|
| 1281 |
+
"loss": 0.6335,
|
| 1282 |
+
"step": 1800
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 0.11584,
|
| 1286 |
+
"grad_norm": 1.9624851942062378,
|
| 1287 |
+
"learning_rate": 7.719223383827608e-06,
|
| 1288 |
+
"loss": 0.6448,
|
| 1289 |
+
"step": 1810
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 0.11648,
|
| 1293 |
+
"grad_norm": 1.4034361839294434,
|
| 1294 |
+
"learning_rate": 7.76189460209089e-06,
|
| 1295 |
+
"loss": 0.6711,
|
| 1296 |
+
"step": 1820
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 0.11712,
|
| 1300 |
+
"grad_norm": 1.503132700920105,
|
| 1301 |
+
"learning_rate": 7.804565820354171e-06,
|
| 1302 |
+
"loss": 0.6457,
|
| 1303 |
+
"step": 1830
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 0.11776,
|
| 1307 |
+
"grad_norm": 1.0341029167175293,
|
| 1308 |
+
"learning_rate": 7.847237038617453e-06,
|
| 1309 |
+
"loss": 0.6393,
|
| 1310 |
+
"step": 1840
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 0.1184,
|
| 1314 |
+
"grad_norm": 1.884297251701355,
|
| 1315 |
+
"learning_rate": 7.889908256880735e-06,
|
| 1316 |
+
"loss": 0.6415,
|
| 1317 |
+
"step": 1850
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 0.11904,
|
| 1321 |
+
"grad_norm": 1.3774992227554321,
|
| 1322 |
+
"learning_rate": 7.932579475144016e-06,
|
| 1323 |
+
"loss": 0.635,
|
| 1324 |
+
"step": 1860
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 0.11968,
|
| 1328 |
+
"grad_norm": 1.9972243309020996,
|
| 1329 |
+
"learning_rate": 7.975250693407298e-06,
|
| 1330 |
+
"loss": 0.64,
|
| 1331 |
+
"step": 1870
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 0.12032,
|
| 1335 |
+
"grad_norm": 1.1215760707855225,
|
| 1336 |
+
"learning_rate": 8.01792191167058e-06,
|
| 1337 |
+
"loss": 0.6358,
|
| 1338 |
+
"step": 1880
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 0.12096,
|
| 1342 |
+
"grad_norm": 1.1372530460357666,
|
| 1343 |
+
"learning_rate": 8.060593129933861e-06,
|
| 1344 |
+
"loss": 0.6264,
|
| 1345 |
+
"step": 1890
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 0.1216,
|
| 1349 |
+
"grad_norm": 1.0789086818695068,
|
| 1350 |
+
"learning_rate": 8.103264348197143e-06,
|
| 1351 |
+
"loss": 0.6396,
|
| 1352 |
+
"step": 1900
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 0.12224,
|
| 1356 |
+
"grad_norm": 1.3709907531738281,
|
| 1357 |
+
"learning_rate": 8.145935566460424e-06,
|
| 1358 |
+
"loss": 0.638,
|
| 1359 |
+
"step": 1910
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 0.12288,
|
| 1363 |
+
"grad_norm": 1.5928348302841187,
|
| 1364 |
+
"learning_rate": 8.188606784723706e-06,
|
| 1365 |
+
"loss": 0.6495,
|
| 1366 |
+
"step": 1920
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 0.12352,
|
| 1370 |
+
"grad_norm": 1.0440688133239746,
|
| 1371 |
+
"learning_rate": 8.231278002986986e-06,
|
| 1372 |
+
"loss": 0.6236,
|
| 1373 |
+
"step": 1930
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 0.12416,
|
| 1377 |
+
"grad_norm": 1.2079739570617676,
|
| 1378 |
+
"learning_rate": 8.273949221250267e-06,
|
| 1379 |
+
"loss": 0.6489,
|
| 1380 |
+
"step": 1940
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 0.1248,
|
| 1384 |
+
"grad_norm": 1.161354422569275,
|
| 1385 |
+
"learning_rate": 8.316620439513549e-06,
|
| 1386 |
+
"loss": 0.6412,
|
| 1387 |
+
"step": 1950
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 0.12544,
|
| 1391 |
+
"grad_norm": 1.7108139991760254,
|
| 1392 |
+
"learning_rate": 8.35929165777683e-06,
|
| 1393 |
+
"loss": 0.6383,
|
| 1394 |
+
"step": 1960
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 0.12608,
|
| 1398 |
+
"grad_norm": 1.4153392314910889,
|
| 1399 |
+
"learning_rate": 8.401962876040112e-06,
|
| 1400 |
+
"loss": 0.6482,
|
| 1401 |
+
"step": 1970
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 0.12672,
|
| 1405 |
+
"grad_norm": 1.6085275411605835,
|
| 1406 |
+
"learning_rate": 8.444634094303393e-06,
|
| 1407 |
+
"loss": 0.6488,
|
| 1408 |
+
"step": 1980
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 0.12736,
|
| 1412 |
+
"grad_norm": 1.2328271865844727,
|
| 1413 |
+
"learning_rate": 8.487305312566675e-06,
|
| 1414 |
+
"loss": 0.6466,
|
| 1415 |
+
"step": 1990
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 0.128,
|
| 1419 |
+
"grad_norm": 0.9865418672561646,
|
| 1420 |
+
"learning_rate": 8.529976530829955e-06,
|
| 1421 |
+
"loss": 0.6468,
|
| 1422 |
+
"step": 2000
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 0.128,
|
| 1426 |
+
"eval_accuracy": 0.7972,
|
| 1427 |
+
"eval_loss": 0.578827440738678,
|
| 1428 |
+
"eval_macro_f1": 0.797186028145479,
|
| 1429 |
+
"eval_precision": 0.7981394308396883,
|
| 1430 |
+
"eval_recall": 0.7985143807471076,
|
| 1431 |
+
"eval_runtime": 813.4399,
|
| 1432 |
+
"eval_samples_per_second": 122.935,
|
| 1433 |
+
"eval_steps_per_second": 1.921,
|
| 1434 |
+
"step": 2000
|
| 1435 |
+
}
|
| 1436 |
+
],
|
| 1437 |
+
"logging_steps": 10,
|
| 1438 |
+
"max_steps": 46875,
|
| 1439 |
+
"num_input_tokens_seen": 0,
|
| 1440 |
+
"num_train_epochs": 3,
|
| 1441 |
+
"save_steps": 200,
|
| 1442 |
+
"stateful_callbacks": {
|
| 1443 |
+
"EarlyStoppingCallback": {
|
| 1444 |
+
"args": {
|
| 1445 |
+
"early_stopping_patience": 3,
|
| 1446 |
+
"early_stopping_threshold": 0.0
|
| 1447 |
+
},
|
| 1448 |
+
"attributes": {
|
| 1449 |
+
"early_stopping_patience_counter": 0
|
| 1450 |
+
}
|
| 1451 |
+
},
|
| 1452 |
+
"TrainerControl": {
|
| 1453 |
+
"args": {
|
| 1454 |
+
"should_epoch_stop": false,
|
| 1455 |
+
"should_evaluate": false,
|
| 1456 |
+
"should_log": false,
|
| 1457 |
+
"should_save": true,
|
| 1458 |
+
"should_training_stop": false
|
| 1459 |
+
},
|
| 1460 |
+
"attributes": {}
|
| 1461 |
+
}
|
| 1462 |
+
},
|
| 1463 |
+
"total_flos": 1.68331875442944e+16,
|
| 1464 |
+
"train_batch_size": 32,
|
| 1465 |
+
"trial_name": null,
|
| 1466 |
+
"trial_params": null
|
| 1467 |
+
}
|
graphcodebert-rdrop/checkpoint-2000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a0e6c50c4bb64305fe23a4e428084f9c072417915284e4ed838c0b852405eb80
|
| 3 |
+
size 5841
|
graphcodebert-rdrop/checkpoint-2000/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
graphcodebert-rdrop/checkpoint-2200/config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.2,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"gradient_checkpointing": false,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout_prob": 0.2,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 514,
|
| 18 |
+
"model_type": "roberta",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"output_past": true,
|
| 22 |
+
"pad_token_id": 1,
|
| 23 |
+
"position_embedding_type": "absolute",
|
| 24 |
+
"transformers_version": "4.56.0",
|
| 25 |
+
"type_vocab_size": 1,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 50265
|
| 28 |
+
}
|
graphcodebert-rdrop/checkpoint-2200/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
graphcodebert-rdrop/checkpoint-2200/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:832cb47285e1ba2464701299ede1a0328d817699ffed22da4964178055f4d577
|
| 3 |
+
size 498612824
|
graphcodebert-rdrop/checkpoint-2200/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ec849a49a27e84cd130dec726a63661e09349b0256b3abc68f4452b429f0a0a
|
| 3 |
+
size 4741923
|
graphcodebert-rdrop/checkpoint-2200/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:714bd65b559f547695a9525e0c9afa49231b1b2a0b649650880af07c3819df59
|
| 3 |
+
size 14645
|
graphcodebert-rdrop/checkpoint-2200/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f6829e47549862e022b4eedf087cf81394f448150e958ad6663988076bf8691
|
| 3 |
+
size 1383
|
graphcodebert-rdrop/checkpoint-2200/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94d198f1873092205d4c42364c530ce00d3f5ca71ab6755f6983b20d7503713e
|
| 3 |
+
size 1465
|
graphcodebert-rdrop/checkpoint-2200/special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
graphcodebert-rdrop/checkpoint-2200/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
graphcodebert-rdrop/checkpoint-2200/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 512,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 56 |
+
"trim_offsets": true,
|
| 57 |
+
"unk_token": "<unk>"
|
| 58 |
+
}
|
graphcodebert-rdrop/checkpoint-2200/trainer_state.json
ADDED
|
@@ -0,0 +1,1607 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 2000,
|
| 3 |
+
"best_metric": 0.797186028145479,
|
| 4 |
+
"best_model_checkpoint": "./output_checkpoints/graphcodebert-rdrop/checkpoint-2000",
|
| 5 |
+
"epoch": 0.1408,
|
| 6 |
+
"eval_steps": 1000,
|
| 7 |
+
"global_step": 2200,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00064,
|
| 14 |
+
"grad_norm": 1.8688431978225708,
|
| 15 |
+
"learning_rate": 3.840409643695328e-08,
|
| 16 |
+
"loss": 0.7168,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.00128,
|
| 21 |
+
"grad_norm": 1.835353970527649,
|
| 22 |
+
"learning_rate": 8.10753147002347e-08,
|
| 23 |
+
"loss": 0.7179,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00192,
|
| 28 |
+
"grad_norm": 1.2541388273239136,
|
| 29 |
+
"learning_rate": 1.2374653296351612e-07,
|
| 30 |
+
"loss": 0.7177,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.00256,
|
| 35 |
+
"grad_norm": 1.1381633281707764,
|
| 36 |
+
"learning_rate": 1.6641775122679754e-07,
|
| 37 |
+
"loss": 0.7312,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0032,
|
| 42 |
+
"grad_norm": 1.6313824653625488,
|
| 43 |
+
"learning_rate": 2.0908896949007894e-07,
|
| 44 |
+
"loss": 0.7167,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.00384,
|
| 49 |
+
"grad_norm": 1.092464566230774,
|
| 50 |
+
"learning_rate": 2.517601877533604e-07,
|
| 51 |
+
"loss": 0.7201,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.00448,
|
| 56 |
+
"grad_norm": 1.164099097251892,
|
| 57 |
+
"learning_rate": 2.944314060166418e-07,
|
| 58 |
+
"loss": 0.7195,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.00512,
|
| 63 |
+
"grad_norm": 1.9283920526504517,
|
| 64 |
+
"learning_rate": 3.371026242799232e-07,
|
| 65 |
+
"loss": 0.7223,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.00576,
|
| 70 |
+
"grad_norm": 1.8716129064559937,
|
| 71 |
+
"learning_rate": 3.7977384254320464e-07,
|
| 72 |
+
"loss": 0.7151,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0064,
|
| 77 |
+
"grad_norm": 1.5499262809753418,
|
| 78 |
+
"learning_rate": 4.22445060806486e-07,
|
| 79 |
+
"loss": 0.7211,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.00704,
|
| 84 |
+
"grad_norm": 1.375114917755127,
|
| 85 |
+
"learning_rate": 4.651162790697675e-07,
|
| 86 |
+
"loss": 0.7247,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.00768,
|
| 91 |
+
"grad_norm": 1.384915828704834,
|
| 92 |
+
"learning_rate": 5.077874973330489e-07,
|
| 93 |
+
"loss": 0.7156,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.00832,
|
| 98 |
+
"grad_norm": 1.6376659870147705,
|
| 99 |
+
"learning_rate": 5.504587155963304e-07,
|
| 100 |
+
"loss": 0.7196,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.00896,
|
| 105 |
+
"grad_norm": 1.709489345550537,
|
| 106 |
+
"learning_rate": 5.931299338596117e-07,
|
| 107 |
+
"loss": 0.7215,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0096,
|
| 112 |
+
"grad_norm": 1.350512146949768,
|
| 113 |
+
"learning_rate": 6.358011521228932e-07,
|
| 114 |
+
"loss": 0.71,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.01024,
|
| 119 |
+
"grad_norm": 2.33050537109375,
|
| 120 |
+
"learning_rate": 6.784723703861745e-07,
|
| 121 |
+
"loss": 0.7191,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.01088,
|
| 126 |
+
"grad_norm": 1.0042874813079834,
|
| 127 |
+
"learning_rate": 7.21143588649456e-07,
|
| 128 |
+
"loss": 0.72,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.01152,
|
| 133 |
+
"grad_norm": 1.1835744380950928,
|
| 134 |
+
"learning_rate": 7.638148069127374e-07,
|
| 135 |
+
"loss": 0.7122,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.01216,
|
| 140 |
+
"grad_norm": 1.949506402015686,
|
| 141 |
+
"learning_rate": 8.064860251760189e-07,
|
| 142 |
+
"loss": 0.7091,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.0128,
|
| 147 |
+
"grad_norm": 1.139626383781433,
|
| 148 |
+
"learning_rate": 8.491572434393003e-07,
|
| 149 |
+
"loss": 0.7046,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.01344,
|
| 154 |
+
"grad_norm": 1.7734779119491577,
|
| 155 |
+
"learning_rate": 8.918284617025817e-07,
|
| 156 |
+
"loss": 0.7135,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.01408,
|
| 161 |
+
"grad_norm": 1.3442974090576172,
|
| 162 |
+
"learning_rate": 9.344996799658632e-07,
|
| 163 |
+
"loss": 0.7127,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.01472,
|
| 168 |
+
"grad_norm": 1.6148335933685303,
|
| 169 |
+
"learning_rate": 9.771708982291445e-07,
|
| 170 |
+
"loss": 0.7168,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.01536,
|
| 175 |
+
"grad_norm": 0.9220213294029236,
|
| 176 |
+
"learning_rate": 1.0198421164924258e-06,
|
| 177 |
+
"loss": 0.7109,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.016,
|
| 182 |
+
"grad_norm": 1.5631815195083618,
|
| 183 |
+
"learning_rate": 1.0625133347557074e-06,
|
| 184 |
+
"loss": 0.7149,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.01664,
|
| 189 |
+
"grad_norm": 1.5052822828292847,
|
| 190 |
+
"learning_rate": 1.1051845530189888e-06,
|
| 191 |
+
"loss": 0.7112,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.01728,
|
| 196 |
+
"grad_norm": 1.6714903116226196,
|
| 197 |
+
"learning_rate": 1.1478557712822702e-06,
|
| 198 |
+
"loss": 0.7128,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.01792,
|
| 203 |
+
"grad_norm": 1.6667262315750122,
|
| 204 |
+
"learning_rate": 1.1905269895455517e-06,
|
| 205 |
+
"loss": 0.7091,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.01856,
|
| 210 |
+
"grad_norm": 1.6861907243728638,
|
| 211 |
+
"learning_rate": 1.233198207808833e-06,
|
| 212 |
+
"loss": 0.7078,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0192,
|
| 217 |
+
"grad_norm": 1.6259212493896484,
|
| 218 |
+
"learning_rate": 1.2758694260721145e-06,
|
| 219 |
+
"loss": 0.7087,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.01984,
|
| 224 |
+
"grad_norm": 1.207320213317871,
|
| 225 |
+
"learning_rate": 1.318540644335396e-06,
|
| 226 |
+
"loss": 0.7087,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.02048,
|
| 231 |
+
"grad_norm": 1.087124228477478,
|
| 232 |
+
"learning_rate": 1.3612118625986772e-06,
|
| 233 |
+
"loss": 0.7174,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.02112,
|
| 238 |
+
"grad_norm": 1.5668721199035645,
|
| 239 |
+
"learning_rate": 1.4038830808619588e-06,
|
| 240 |
+
"loss": 0.7066,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.02176,
|
| 245 |
+
"grad_norm": 1.5332859754562378,
|
| 246 |
+
"learning_rate": 1.4465542991252401e-06,
|
| 247 |
+
"loss": 0.7115,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0224,
|
| 252 |
+
"grad_norm": 2.092994451522827,
|
| 253 |
+
"learning_rate": 1.4892255173885215e-06,
|
| 254 |
+
"loss": 0.7085,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.02304,
|
| 259 |
+
"grad_norm": 1.266851544380188,
|
| 260 |
+
"learning_rate": 1.531896735651803e-06,
|
| 261 |
+
"loss": 0.7091,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.02368,
|
| 266 |
+
"grad_norm": 1.290616512298584,
|
| 267 |
+
"learning_rate": 1.5745679539150842e-06,
|
| 268 |
+
"loss": 0.7103,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.02432,
|
| 273 |
+
"grad_norm": 1.2045000791549683,
|
| 274 |
+
"learning_rate": 1.6172391721783658e-06,
|
| 275 |
+
"loss": 0.705,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.02496,
|
| 280 |
+
"grad_norm": 1.828832983970642,
|
| 281 |
+
"learning_rate": 1.6599103904416472e-06,
|
| 282 |
+
"loss": 0.7038,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.0256,
|
| 287 |
+
"grad_norm": 1.7793196439743042,
|
| 288 |
+
"learning_rate": 1.7025816087049288e-06,
|
| 289 |
+
"loss": 0.7085,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.02624,
|
| 294 |
+
"grad_norm": 0.8949472904205322,
|
| 295 |
+
"learning_rate": 1.7452528269682101e-06,
|
| 296 |
+
"loss": 0.7045,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.02688,
|
| 301 |
+
"grad_norm": 1.3071945905685425,
|
| 302 |
+
"learning_rate": 1.7879240452314913e-06,
|
| 303 |
+
"loss": 0.702,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.02752,
|
| 308 |
+
"grad_norm": 1.7497148513793945,
|
| 309 |
+
"learning_rate": 1.8305952634947729e-06,
|
| 310 |
+
"loss": 0.7073,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.02816,
|
| 315 |
+
"grad_norm": 1.0856297016143799,
|
| 316 |
+
"learning_rate": 1.8732664817580542e-06,
|
| 317 |
+
"loss": 0.7008,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.0288,
|
| 322 |
+
"grad_norm": 1.369019627571106,
|
| 323 |
+
"learning_rate": 1.915937700021336e-06,
|
| 324 |
+
"loss": 0.7059,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.02944,
|
| 329 |
+
"grad_norm": 2.0961010456085205,
|
| 330 |
+
"learning_rate": 1.958608918284617e-06,
|
| 331 |
+
"loss": 0.6952,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.03008,
|
| 336 |
+
"grad_norm": 1.6076347827911377,
|
| 337 |
+
"learning_rate": 2.0012801365478988e-06,
|
| 338 |
+
"loss": 0.7074,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.03072,
|
| 343 |
+
"grad_norm": 1.3129311800003052,
|
| 344 |
+
"learning_rate": 2.04395135481118e-06,
|
| 345 |
+
"loss": 0.6994,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.03136,
|
| 350 |
+
"grad_norm": 1.3621476888656616,
|
| 351 |
+
"learning_rate": 2.0866225730744615e-06,
|
| 352 |
+
"loss": 0.696,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.032,
|
| 357 |
+
"grad_norm": 0.9786806106567383,
|
| 358 |
+
"learning_rate": 2.129293791337743e-06,
|
| 359 |
+
"loss": 0.7011,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.03264,
|
| 364 |
+
"grad_norm": 1.4611176252365112,
|
| 365 |
+
"learning_rate": 2.1719650096010242e-06,
|
| 366 |
+
"loss": 0.699,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.03328,
|
| 371 |
+
"grad_norm": 1.0675945281982422,
|
| 372 |
+
"learning_rate": 2.214636227864306e-06,
|
| 373 |
+
"loss": 0.6906,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.03392,
|
| 378 |
+
"grad_norm": 1.6564017534255981,
|
| 379 |
+
"learning_rate": 2.257307446127587e-06,
|
| 380 |
+
"loss": 0.6923,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.03456,
|
| 385 |
+
"grad_norm": 1.228119134902954,
|
| 386 |
+
"learning_rate": 2.2999786643908685e-06,
|
| 387 |
+
"loss": 0.6928,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.0352,
|
| 392 |
+
"grad_norm": 1.9868593215942383,
|
| 393 |
+
"learning_rate": 2.34264988265415e-06,
|
| 394 |
+
"loss": 0.6912,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.03584,
|
| 399 |
+
"grad_norm": 1.2531176805496216,
|
| 400 |
+
"learning_rate": 2.3853211009174317e-06,
|
| 401 |
+
"loss": 0.6923,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.03648,
|
| 406 |
+
"grad_norm": 1.413602352142334,
|
| 407 |
+
"learning_rate": 2.427992319180713e-06,
|
| 408 |
+
"loss": 0.6993,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.03712,
|
| 413 |
+
"grad_norm": 0.9655390977859497,
|
| 414 |
+
"learning_rate": 2.470663537443994e-06,
|
| 415 |
+
"loss": 0.7017,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.03776,
|
| 420 |
+
"grad_norm": 1.962438941001892,
|
| 421 |
+
"learning_rate": 2.5133347557072756e-06,
|
| 422 |
+
"loss": 0.6933,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.0384,
|
| 427 |
+
"grad_norm": 1.1099931001663208,
|
| 428 |
+
"learning_rate": 2.556005973970557e-06,
|
| 429 |
+
"loss": 0.6925,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.03904,
|
| 434 |
+
"grad_norm": 1.1766624450683594,
|
| 435 |
+
"learning_rate": 2.5986771922338383e-06,
|
| 436 |
+
"loss": 0.6894,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.03968,
|
| 441 |
+
"grad_norm": 1.574353814125061,
|
| 442 |
+
"learning_rate": 2.64134841049712e-06,
|
| 443 |
+
"loss": 0.6923,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.04032,
|
| 448 |
+
"grad_norm": 1.194074273109436,
|
| 449 |
+
"learning_rate": 2.6840196287604015e-06,
|
| 450 |
+
"loss": 0.6855,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.04096,
|
| 455 |
+
"grad_norm": 1.1750593185424805,
|
| 456 |
+
"learning_rate": 2.7266908470236826e-06,
|
| 457 |
+
"loss": 0.7002,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.0416,
|
| 462 |
+
"grad_norm": 1.3165347576141357,
|
| 463 |
+
"learning_rate": 2.7693620652869642e-06,
|
| 464 |
+
"loss": 0.6893,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.04224,
|
| 469 |
+
"grad_norm": 1.1079384088516235,
|
| 470 |
+
"learning_rate": 2.812033283550246e-06,
|
| 471 |
+
"loss": 0.6852,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.04288,
|
| 476 |
+
"grad_norm": 1.231327772140503,
|
| 477 |
+
"learning_rate": 2.8547045018135274e-06,
|
| 478 |
+
"loss": 0.695,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.04352,
|
| 483 |
+
"grad_norm": 1.966036319732666,
|
| 484 |
+
"learning_rate": 2.897375720076808e-06,
|
| 485 |
+
"loss": 0.6833,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.04416,
|
| 490 |
+
"grad_norm": 1.0460131168365479,
|
| 491 |
+
"learning_rate": 2.9400469383400897e-06,
|
| 492 |
+
"loss": 0.6802,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.0448,
|
| 497 |
+
"grad_norm": 1.2445200681686401,
|
| 498 |
+
"learning_rate": 2.9827181566033713e-06,
|
| 499 |
+
"loss": 0.6883,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.04544,
|
| 504 |
+
"grad_norm": 1.1759636402130127,
|
| 505 |
+
"learning_rate": 3.025389374866653e-06,
|
| 506 |
+
"loss": 0.682,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.04608,
|
| 511 |
+
"grad_norm": 1.2478561401367188,
|
| 512 |
+
"learning_rate": 3.068060593129934e-06,
|
| 513 |
+
"loss": 0.6802,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.04672,
|
| 518 |
+
"grad_norm": 0.8945108652114868,
|
| 519 |
+
"learning_rate": 3.1107318113932156e-06,
|
| 520 |
+
"loss": 0.6873,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.04736,
|
| 525 |
+
"grad_norm": 1.6810317039489746,
|
| 526 |
+
"learning_rate": 3.153403029656497e-06,
|
| 527 |
+
"loss": 0.6828,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.048,
|
| 532 |
+
"grad_norm": 2.882283926010132,
|
| 533 |
+
"learning_rate": 3.1960742479197783e-06,
|
| 534 |
+
"loss": 0.6797,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.04864,
|
| 539 |
+
"grad_norm": 1.575766921043396,
|
| 540 |
+
"learning_rate": 3.23874546618306e-06,
|
| 541 |
+
"loss": 0.6826,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.04928,
|
| 546 |
+
"grad_norm": 1.8044737577438354,
|
| 547 |
+
"learning_rate": 3.2814166844463415e-06,
|
| 548 |
+
"loss": 0.6837,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.04992,
|
| 553 |
+
"grad_norm": 1.4755513668060303,
|
| 554 |
+
"learning_rate": 3.324087902709623e-06,
|
| 555 |
+
"loss": 0.6815,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.05056,
|
| 560 |
+
"grad_norm": 2.0735654830932617,
|
| 561 |
+
"learning_rate": 3.3667591209729038e-06,
|
| 562 |
+
"loss": 0.6773,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.0512,
|
| 567 |
+
"grad_norm": 0.9823655486106873,
|
| 568 |
+
"learning_rate": 3.4094303392361854e-06,
|
| 569 |
+
"loss": 0.6689,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.05184,
|
| 574 |
+
"grad_norm": 1.0731920003890991,
|
| 575 |
+
"learning_rate": 3.452101557499467e-06,
|
| 576 |
+
"loss": 0.6689,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.05248,
|
| 581 |
+
"grad_norm": 1.1081087589263916,
|
| 582 |
+
"learning_rate": 3.494772775762748e-06,
|
| 583 |
+
"loss": 0.6694,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.05312,
|
| 588 |
+
"grad_norm": 1.2130361795425415,
|
| 589 |
+
"learning_rate": 3.5374439940260297e-06,
|
| 590 |
+
"loss": 0.682,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.05376,
|
| 595 |
+
"grad_norm": 1.170345664024353,
|
| 596 |
+
"learning_rate": 3.5801152122893113e-06,
|
| 597 |
+
"loss": 0.6756,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.0544,
|
| 602 |
+
"grad_norm": 1.2610124349594116,
|
| 603 |
+
"learning_rate": 3.622786430552593e-06,
|
| 604 |
+
"loss": 0.67,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.05504,
|
| 609 |
+
"grad_norm": 1.0138545036315918,
|
| 610 |
+
"learning_rate": 3.665457648815874e-06,
|
| 611 |
+
"loss": 0.6807,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.05568,
|
| 616 |
+
"grad_norm": 1.0937036275863647,
|
| 617 |
+
"learning_rate": 3.7081288670791556e-06,
|
| 618 |
+
"loss": 0.6774,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.05632,
|
| 623 |
+
"grad_norm": 1.119221568107605,
|
| 624 |
+
"learning_rate": 3.750800085342437e-06,
|
| 625 |
+
"loss": 0.6728,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.05696,
|
| 630 |
+
"grad_norm": 1.9351646900177002,
|
| 631 |
+
"learning_rate": 3.793471303605718e-06,
|
| 632 |
+
"loss": 0.6664,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.0576,
|
| 637 |
+
"grad_norm": 1.6208665370941162,
|
| 638 |
+
"learning_rate": 3.836142521869e-06,
|
| 639 |
+
"loss": 0.6653,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.05824,
|
| 644 |
+
"grad_norm": 1.3460793495178223,
|
| 645 |
+
"learning_rate": 3.8788137401322815e-06,
|
| 646 |
+
"loss": 0.6651,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.05888,
|
| 651 |
+
"grad_norm": 1.8987629413604736,
|
| 652 |
+
"learning_rate": 3.921484958395563e-06,
|
| 653 |
+
"loss": 0.6694,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.05952,
|
| 658 |
+
"grad_norm": 1.3637055158615112,
|
| 659 |
+
"learning_rate": 3.964156176658844e-06,
|
| 660 |
+
"loss": 0.6691,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.06016,
|
| 665 |
+
"grad_norm": 1.3319580554962158,
|
| 666 |
+
"learning_rate": 4.006827394922125e-06,
|
| 667 |
+
"loss": 0.6749,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.0608,
|
| 672 |
+
"grad_norm": 1.3574756383895874,
|
| 673 |
+
"learning_rate": 4.049498613185407e-06,
|
| 674 |
+
"loss": 0.6688,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.06144,
|
| 679 |
+
"grad_norm": 1.136319637298584,
|
| 680 |
+
"learning_rate": 4.092169831448688e-06,
|
| 681 |
+
"loss": 0.6657,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.06208,
|
| 686 |
+
"grad_norm": 1.7666376829147339,
|
| 687 |
+
"learning_rate": 4.134841049711969e-06,
|
| 688 |
+
"loss": 0.6667,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.06272,
|
| 693 |
+
"grad_norm": 1.332014560699463,
|
| 694 |
+
"learning_rate": 4.177512267975251e-06,
|
| 695 |
+
"loss": 0.6715,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.06336,
|
| 700 |
+
"grad_norm": 1.3127824068069458,
|
| 701 |
+
"learning_rate": 4.220183486238532e-06,
|
| 702 |
+
"loss": 0.6737,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.064,
|
| 707 |
+
"grad_norm": 1.293678641319275,
|
| 708 |
+
"learning_rate": 4.262854704501814e-06,
|
| 709 |
+
"loss": 0.6557,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.064,
|
| 714 |
+
"eval_accuracy": 0.76589,
|
| 715 |
+
"eval_loss": 0.6282070279121399,
|
| 716 |
+
"eval_macro_f1": 0.7657238614598489,
|
| 717 |
+
"eval_precision": 0.7708542319194018,
|
| 718 |
+
"eval_recall": 0.7687514011652705,
|
| 719 |
+
"eval_runtime": 806.5969,
|
| 720 |
+
"eval_samples_per_second": 123.978,
|
| 721 |
+
"eval_steps_per_second": 1.938,
|
| 722 |
+
"step": 1000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.06464,
|
| 726 |
+
"grad_norm": 1.2433074712753296,
|
| 727 |
+
"learning_rate": 4.3055259227650956e-06,
|
| 728 |
+
"loss": 0.6583,
|
| 729 |
+
"step": 1010
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.06528,
|
| 733 |
+
"grad_norm": 1.5115686655044556,
|
| 734 |
+
"learning_rate": 4.348197141028377e-06,
|
| 735 |
+
"loss": 0.6659,
|
| 736 |
+
"step": 1020
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 0.06592,
|
| 740 |
+
"grad_norm": 1.3560088872909546,
|
| 741 |
+
"learning_rate": 4.390868359291658e-06,
|
| 742 |
+
"loss": 0.6675,
|
| 743 |
+
"step": 1030
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 0.06656,
|
| 747 |
+
"grad_norm": 0.8229928016662598,
|
| 748 |
+
"learning_rate": 4.4335395775549394e-06,
|
| 749 |
+
"loss": 0.6629,
|
| 750 |
+
"step": 1040
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 0.0672,
|
| 754 |
+
"grad_norm": 1.5138262510299683,
|
| 755 |
+
"learning_rate": 4.476210795818221e-06,
|
| 756 |
+
"loss": 0.6635,
|
| 757 |
+
"step": 1050
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 0.06784,
|
| 761 |
+
"grad_norm": 1.1732233762741089,
|
| 762 |
+
"learning_rate": 4.518882014081503e-06,
|
| 763 |
+
"loss": 0.6531,
|
| 764 |
+
"step": 1060
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.06848,
|
| 768 |
+
"grad_norm": 1.6118066310882568,
|
| 769 |
+
"learning_rate": 4.561553232344783e-06,
|
| 770 |
+
"loss": 0.6644,
|
| 771 |
+
"step": 1070
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 0.06912,
|
| 775 |
+
"grad_norm": 0.9575800895690918,
|
| 776 |
+
"learning_rate": 4.604224450608065e-06,
|
| 777 |
+
"loss": 0.6542,
|
| 778 |
+
"step": 1080
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.06976,
|
| 782 |
+
"grad_norm": 1.157271146774292,
|
| 783 |
+
"learning_rate": 4.6468956688713465e-06,
|
| 784 |
+
"loss": 0.6505,
|
| 785 |
+
"step": 1090
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.0704,
|
| 789 |
+
"grad_norm": 1.4502589702606201,
|
| 790 |
+
"learning_rate": 4.689566887134628e-06,
|
| 791 |
+
"loss": 0.6725,
|
| 792 |
+
"step": 1100
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 0.07104,
|
| 796 |
+
"grad_norm": 1.122251033782959,
|
| 797 |
+
"learning_rate": 4.73223810539791e-06,
|
| 798 |
+
"loss": 0.6625,
|
| 799 |
+
"step": 1110
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 0.07168,
|
| 803 |
+
"grad_norm": 0.9529566764831543,
|
| 804 |
+
"learning_rate": 4.774909323661191e-06,
|
| 805 |
+
"loss": 0.6684,
|
| 806 |
+
"step": 1120
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.07232,
|
| 810 |
+
"grad_norm": 1.6433424949645996,
|
| 811 |
+
"learning_rate": 4.817580541924473e-06,
|
| 812 |
+
"loss": 0.6599,
|
| 813 |
+
"step": 1130
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.07296,
|
| 817 |
+
"grad_norm": 1.6565576791763306,
|
| 818 |
+
"learning_rate": 4.8602517601877535e-06,
|
| 819 |
+
"loss": 0.6591,
|
| 820 |
+
"step": 1140
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 0.0736,
|
| 824 |
+
"grad_norm": 1.3955042362213135,
|
| 825 |
+
"learning_rate": 4.902922978451035e-06,
|
| 826 |
+
"loss": 0.6649,
|
| 827 |
+
"step": 1150
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 0.07424,
|
| 831 |
+
"grad_norm": 1.530819296836853,
|
| 832 |
+
"learning_rate": 4.945594196714317e-06,
|
| 833 |
+
"loss": 0.65,
|
| 834 |
+
"step": 1160
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.07488,
|
| 838 |
+
"grad_norm": 1.0729092359542847,
|
| 839 |
+
"learning_rate": 4.9882654149775974e-06,
|
| 840 |
+
"loss": 0.6603,
|
| 841 |
+
"step": 1170
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 0.07552,
|
| 845 |
+
"grad_norm": 1.6307648420333862,
|
| 846 |
+
"learning_rate": 5.030936633240879e-06,
|
| 847 |
+
"loss": 0.6702,
|
| 848 |
+
"step": 1180
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.07616,
|
| 852 |
+
"grad_norm": 1.324406623840332,
|
| 853 |
+
"learning_rate": 5.073607851504161e-06,
|
| 854 |
+
"loss": 0.6634,
|
| 855 |
+
"step": 1190
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 0.0768,
|
| 859 |
+
"grad_norm": 1.1919140815734863,
|
| 860 |
+
"learning_rate": 5.116279069767442e-06,
|
| 861 |
+
"loss": 0.6586,
|
| 862 |
+
"step": 1200
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 0.07744,
|
| 866 |
+
"grad_norm": 1.0658751726150513,
|
| 867 |
+
"learning_rate": 5.158950288030724e-06,
|
| 868 |
+
"loss": 0.6573,
|
| 869 |
+
"step": 1210
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 0.07808,
|
| 873 |
+
"grad_norm": 1.7762951850891113,
|
| 874 |
+
"learning_rate": 5.201621506294005e-06,
|
| 875 |
+
"loss": 0.6596,
|
| 876 |
+
"step": 1220
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 0.07872,
|
| 880 |
+
"grad_norm": 1.1911667585372925,
|
| 881 |
+
"learning_rate": 5.244292724557286e-06,
|
| 882 |
+
"loss": 0.651,
|
| 883 |
+
"step": 1230
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 0.07936,
|
| 887 |
+
"grad_norm": 1.442029595375061,
|
| 888 |
+
"learning_rate": 5.286963942820568e-06,
|
| 889 |
+
"loss": 0.6598,
|
| 890 |
+
"step": 1240
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.08,
|
| 894 |
+
"grad_norm": 0.9955100417137146,
|
| 895 |
+
"learning_rate": 5.329635161083849e-06,
|
| 896 |
+
"loss": 0.6587,
|
| 897 |
+
"step": 1250
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.08064,
|
| 901 |
+
"grad_norm": 1.1801563501358032,
|
| 902 |
+
"learning_rate": 5.372306379347131e-06,
|
| 903 |
+
"loss": 0.6669,
|
| 904 |
+
"step": 1260
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 0.08128,
|
| 908 |
+
"grad_norm": 1.8548506498336792,
|
| 909 |
+
"learning_rate": 5.414977597610412e-06,
|
| 910 |
+
"loss": 0.6534,
|
| 911 |
+
"step": 1270
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.08192,
|
| 915 |
+
"grad_norm": 1.8297194242477417,
|
| 916 |
+
"learning_rate": 5.457648815873694e-06,
|
| 917 |
+
"loss": 0.668,
|
| 918 |
+
"step": 1280
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 0.08256,
|
| 922 |
+
"grad_norm": 1.3471113443374634,
|
| 923 |
+
"learning_rate": 5.5003200341369755e-06,
|
| 924 |
+
"loss": 0.659,
|
| 925 |
+
"step": 1290
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 0.0832,
|
| 929 |
+
"grad_norm": 1.4015250205993652,
|
| 930 |
+
"learning_rate": 5.542991252400256e-06,
|
| 931 |
+
"loss": 0.6602,
|
| 932 |
+
"step": 1300
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.08384,
|
| 936 |
+
"grad_norm": 1.0796362161636353,
|
| 937 |
+
"learning_rate": 5.585662470663538e-06,
|
| 938 |
+
"loss": 0.6652,
|
| 939 |
+
"step": 1310
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.08448,
|
| 943 |
+
"grad_norm": 2.0007293224334717,
|
| 944 |
+
"learning_rate": 5.628333688926819e-06,
|
| 945 |
+
"loss": 0.6573,
|
| 946 |
+
"step": 1320
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.08512,
|
| 950 |
+
"grad_norm": 1.3140546083450317,
|
| 951 |
+
"learning_rate": 5.671004907190101e-06,
|
| 952 |
+
"loss": 0.6642,
|
| 953 |
+
"step": 1330
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 0.08576,
|
| 957 |
+
"grad_norm": 1.6801965236663818,
|
| 958 |
+
"learning_rate": 5.7136761254533826e-06,
|
| 959 |
+
"loss": 0.6527,
|
| 960 |
+
"step": 1340
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 0.0864,
|
| 964 |
+
"grad_norm": 1.9289913177490234,
|
| 965 |
+
"learning_rate": 5.756347343716664e-06,
|
| 966 |
+
"loss": 0.6567,
|
| 967 |
+
"step": 1350
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 0.08704,
|
| 971 |
+
"grad_norm": 1.420436143875122,
|
| 972 |
+
"learning_rate": 5.799018561979946e-06,
|
| 973 |
+
"loss": 0.6479,
|
| 974 |
+
"step": 1360
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.08768,
|
| 978 |
+
"grad_norm": 1.3068914413452148,
|
| 979 |
+
"learning_rate": 5.841689780243226e-06,
|
| 980 |
+
"loss": 0.6522,
|
| 981 |
+
"step": 1370
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 0.08832,
|
| 985 |
+
"grad_norm": 1.8793392181396484,
|
| 986 |
+
"learning_rate": 5.884360998506507e-06,
|
| 987 |
+
"loss": 0.6491,
|
| 988 |
+
"step": 1380
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 0.08896,
|
| 992 |
+
"grad_norm": 1.3998247385025024,
|
| 993 |
+
"learning_rate": 5.927032216769789e-06,
|
| 994 |
+
"loss": 0.6638,
|
| 995 |
+
"step": 1390
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 0.0896,
|
| 999 |
+
"grad_norm": 1.371824026107788,
|
| 1000 |
+
"learning_rate": 5.96970343503307e-06,
|
| 1001 |
+
"loss": 0.6517,
|
| 1002 |
+
"step": 1400
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.09024,
|
| 1006 |
+
"grad_norm": 2.0050251483917236,
|
| 1007 |
+
"learning_rate": 6.012374653296352e-06,
|
| 1008 |
+
"loss": 0.6556,
|
| 1009 |
+
"step": 1410
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.09088,
|
| 1013 |
+
"grad_norm": 1.499855399131775,
|
| 1014 |
+
"learning_rate": 6.0550458715596335e-06,
|
| 1015 |
+
"loss": 0.6597,
|
| 1016 |
+
"step": 1420
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.09152,
|
| 1020 |
+
"grad_norm": 1.2066516876220703,
|
| 1021 |
+
"learning_rate": 6.097717089822915e-06,
|
| 1022 |
+
"loss": 0.6546,
|
| 1023 |
+
"step": 1430
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.09216,
|
| 1027 |
+
"grad_norm": 1.3586395978927612,
|
| 1028 |
+
"learning_rate": 6.140388308086196e-06,
|
| 1029 |
+
"loss": 0.6434,
|
| 1030 |
+
"step": 1440
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 0.0928,
|
| 1034 |
+
"grad_norm": 1.1413036584854126,
|
| 1035 |
+
"learning_rate": 6.183059526349477e-06,
|
| 1036 |
+
"loss": 0.6554,
|
| 1037 |
+
"step": 1450
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.09344,
|
| 1041 |
+
"grad_norm": 1.1844472885131836,
|
| 1042 |
+
"learning_rate": 6.225730744612759e-06,
|
| 1043 |
+
"loss": 0.6357,
|
| 1044 |
+
"step": 1460
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.09408,
|
| 1048 |
+
"grad_norm": 1.7203835248947144,
|
| 1049 |
+
"learning_rate": 6.2684019628760406e-06,
|
| 1050 |
+
"loss": 0.6546,
|
| 1051 |
+
"step": 1470
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.09472,
|
| 1055 |
+
"grad_norm": 1.1869184970855713,
|
| 1056 |
+
"learning_rate": 6.311073181139322e-06,
|
| 1057 |
+
"loss": 0.6376,
|
| 1058 |
+
"step": 1480
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.09536,
|
| 1062 |
+
"grad_norm": 1.3234835863113403,
|
| 1063 |
+
"learning_rate": 6.353744399402604e-06,
|
| 1064 |
+
"loss": 0.6494,
|
| 1065 |
+
"step": 1490
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.096,
|
| 1069 |
+
"grad_norm": 1.1924010515213013,
|
| 1070 |
+
"learning_rate": 6.396415617665885e-06,
|
| 1071 |
+
"loss": 0.652,
|
| 1072 |
+
"step": 1500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.09664,
|
| 1076 |
+
"grad_norm": 1.4505853652954102,
|
| 1077 |
+
"learning_rate": 6.439086835929166e-06,
|
| 1078 |
+
"loss": 0.6606,
|
| 1079 |
+
"step": 1510
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.09728,
|
| 1083 |
+
"grad_norm": 1.6351200342178345,
|
| 1084 |
+
"learning_rate": 6.481758054192448e-06,
|
| 1085 |
+
"loss": 0.6567,
|
| 1086 |
+
"step": 1520
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.09792,
|
| 1090 |
+
"grad_norm": 1.4423909187316895,
|
| 1091 |
+
"learning_rate": 6.524429272455729e-06,
|
| 1092 |
+
"loss": 0.653,
|
| 1093 |
+
"step": 1530
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 0.09856,
|
| 1097 |
+
"grad_norm": 0.8749169111251831,
|
| 1098 |
+
"learning_rate": 6.567100490719011e-06,
|
| 1099 |
+
"loss": 0.6417,
|
| 1100 |
+
"step": 1540
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 0.0992,
|
| 1104 |
+
"grad_norm": 1.2707208395004272,
|
| 1105 |
+
"learning_rate": 6.609771708982292e-06,
|
| 1106 |
+
"loss": 0.6563,
|
| 1107 |
+
"step": 1550
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 0.09984,
|
| 1111 |
+
"grad_norm": 1.163254976272583,
|
| 1112 |
+
"learning_rate": 6.652442927245574e-06,
|
| 1113 |
+
"loss": 0.6425,
|
| 1114 |
+
"step": 1560
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 0.10048,
|
| 1118 |
+
"grad_norm": 1.4674128293991089,
|
| 1119 |
+
"learning_rate": 6.6951141455088555e-06,
|
| 1120 |
+
"loss": 0.6561,
|
| 1121 |
+
"step": 1570
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 0.10112,
|
| 1125 |
+
"grad_norm": 1.6261919736862183,
|
| 1126 |
+
"learning_rate": 6.737785363772135e-06,
|
| 1127 |
+
"loss": 0.6473,
|
| 1128 |
+
"step": 1580
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 0.10176,
|
| 1132 |
+
"grad_norm": 1.2146090269088745,
|
| 1133 |
+
"learning_rate": 6.780456582035417e-06,
|
| 1134 |
+
"loss": 0.6467,
|
| 1135 |
+
"step": 1590
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 0.1024,
|
| 1139 |
+
"grad_norm": 1.3596712350845337,
|
| 1140 |
+
"learning_rate": 6.8231278002986985e-06,
|
| 1141 |
+
"loss": 0.6431,
|
| 1142 |
+
"step": 1600
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 0.10304,
|
| 1146 |
+
"grad_norm": 1.7861741781234741,
|
| 1147 |
+
"learning_rate": 6.86579901856198e-06,
|
| 1148 |
+
"loss": 0.6586,
|
| 1149 |
+
"step": 1610
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.10368,
|
| 1153 |
+
"grad_norm": 1.456587553024292,
|
| 1154 |
+
"learning_rate": 6.908470236825262e-06,
|
| 1155 |
+
"loss": 0.6496,
|
| 1156 |
+
"step": 1620
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 0.10432,
|
| 1160 |
+
"grad_norm": 1.5652499198913574,
|
| 1161 |
+
"learning_rate": 6.951141455088543e-06,
|
| 1162 |
+
"loss": 0.6436,
|
| 1163 |
+
"step": 1630
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 0.10496,
|
| 1167 |
+
"grad_norm": 1.1411017179489136,
|
| 1168 |
+
"learning_rate": 6.993812673351825e-06,
|
| 1169 |
+
"loss": 0.6572,
|
| 1170 |
+
"step": 1640
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 0.1056,
|
| 1174 |
+
"grad_norm": 1.3867056369781494,
|
| 1175 |
+
"learning_rate": 7.036483891615106e-06,
|
| 1176 |
+
"loss": 0.6425,
|
| 1177 |
+
"step": 1650
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 0.10624,
|
| 1181 |
+
"grad_norm": 1.2714463472366333,
|
| 1182 |
+
"learning_rate": 7.079155109878387e-06,
|
| 1183 |
+
"loss": 0.6558,
|
| 1184 |
+
"step": 1660
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 0.10688,
|
| 1188 |
+
"grad_norm": 1.32753586769104,
|
| 1189 |
+
"learning_rate": 7.121826328141669e-06,
|
| 1190 |
+
"loss": 0.6537,
|
| 1191 |
+
"step": 1670
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 0.10752,
|
| 1195 |
+
"grad_norm": 1.5559014081954956,
|
| 1196 |
+
"learning_rate": 7.16449754640495e-06,
|
| 1197 |
+
"loss": 0.6613,
|
| 1198 |
+
"step": 1680
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 0.10816,
|
| 1202 |
+
"grad_norm": 1.1726677417755127,
|
| 1203 |
+
"learning_rate": 7.207168764668232e-06,
|
| 1204 |
+
"loss": 0.6516,
|
| 1205 |
+
"step": 1690
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 0.1088,
|
| 1209 |
+
"grad_norm": 1.3453316688537598,
|
| 1210 |
+
"learning_rate": 7.2498399829315135e-06,
|
| 1211 |
+
"loss": 0.6455,
|
| 1212 |
+
"step": 1700
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 0.10944,
|
| 1216 |
+
"grad_norm": 1.6420834064483643,
|
| 1217 |
+
"learning_rate": 7.292511201194795e-06,
|
| 1218 |
+
"loss": 0.6457,
|
| 1219 |
+
"step": 1710
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 0.11008,
|
| 1223 |
+
"grad_norm": 1.0518358945846558,
|
| 1224 |
+
"learning_rate": 7.335182419458076e-06,
|
| 1225 |
+
"loss": 0.6315,
|
| 1226 |
+
"step": 1720
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 0.11072,
|
| 1230 |
+
"grad_norm": 1.4042829275131226,
|
| 1231 |
+
"learning_rate": 7.377853637721357e-06,
|
| 1232 |
+
"loss": 0.6445,
|
| 1233 |
+
"step": 1730
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 0.11136,
|
| 1237 |
+
"grad_norm": 1.2282599210739136,
|
| 1238 |
+
"learning_rate": 7.420524855984639e-06,
|
| 1239 |
+
"loss": 0.648,
|
| 1240 |
+
"step": 1740
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 0.112,
|
| 1244 |
+
"grad_norm": 1.2711114883422852,
|
| 1245 |
+
"learning_rate": 7.4631960742479205e-06,
|
| 1246 |
+
"loss": 0.6522,
|
| 1247 |
+
"step": 1750
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 0.11264,
|
| 1251 |
+
"grad_norm": 1.2256455421447754,
|
| 1252 |
+
"learning_rate": 7.505867292511202e-06,
|
| 1253 |
+
"loss": 0.6414,
|
| 1254 |
+
"step": 1760
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 0.11328,
|
| 1258 |
+
"grad_norm": 1.2845845222473145,
|
| 1259 |
+
"learning_rate": 7.548538510774484e-06,
|
| 1260 |
+
"loss": 0.6372,
|
| 1261 |
+
"step": 1770
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 0.11392,
|
| 1265 |
+
"grad_norm": 1.557308554649353,
|
| 1266 |
+
"learning_rate": 7.591209729037765e-06,
|
| 1267 |
+
"loss": 0.6472,
|
| 1268 |
+
"step": 1780
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 0.11456,
|
| 1272 |
+
"grad_norm": 1.5465153455734253,
|
| 1273 |
+
"learning_rate": 7.633880947301045e-06,
|
| 1274 |
+
"loss": 0.6387,
|
| 1275 |
+
"step": 1790
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 0.1152,
|
| 1279 |
+
"grad_norm": 1.0359770059585571,
|
| 1280 |
+
"learning_rate": 7.676552165564327e-06,
|
| 1281 |
+
"loss": 0.6335,
|
| 1282 |
+
"step": 1800
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 0.11584,
|
| 1286 |
+
"grad_norm": 1.9624851942062378,
|
| 1287 |
+
"learning_rate": 7.719223383827608e-06,
|
| 1288 |
+
"loss": 0.6448,
|
| 1289 |
+
"step": 1810
|
| 1290 |
+
},
|
| 1291 |
+
{
|
| 1292 |
+
"epoch": 0.11648,
|
| 1293 |
+
"grad_norm": 1.4034361839294434,
|
| 1294 |
+
"learning_rate": 7.76189460209089e-06,
|
| 1295 |
+
"loss": 0.6711,
|
| 1296 |
+
"step": 1820
|
| 1297 |
+
},
|
| 1298 |
+
{
|
| 1299 |
+
"epoch": 0.11712,
|
| 1300 |
+
"grad_norm": 1.503132700920105,
|
| 1301 |
+
"learning_rate": 7.804565820354171e-06,
|
| 1302 |
+
"loss": 0.6457,
|
| 1303 |
+
"step": 1830
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 0.11776,
|
| 1307 |
+
"grad_norm": 1.0341029167175293,
|
| 1308 |
+
"learning_rate": 7.847237038617453e-06,
|
| 1309 |
+
"loss": 0.6393,
|
| 1310 |
+
"step": 1840
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 0.1184,
|
| 1314 |
+
"grad_norm": 1.884297251701355,
|
| 1315 |
+
"learning_rate": 7.889908256880735e-06,
|
| 1316 |
+
"loss": 0.6415,
|
| 1317 |
+
"step": 1850
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 0.11904,
|
| 1321 |
+
"grad_norm": 1.3774992227554321,
|
| 1322 |
+
"learning_rate": 7.932579475144016e-06,
|
| 1323 |
+
"loss": 0.635,
|
| 1324 |
+
"step": 1860
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 0.11968,
|
| 1328 |
+
"grad_norm": 1.9972243309020996,
|
| 1329 |
+
"learning_rate": 7.975250693407298e-06,
|
| 1330 |
+
"loss": 0.64,
|
| 1331 |
+
"step": 1870
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 0.12032,
|
| 1335 |
+
"grad_norm": 1.1215760707855225,
|
| 1336 |
+
"learning_rate": 8.01792191167058e-06,
|
| 1337 |
+
"loss": 0.6358,
|
| 1338 |
+
"step": 1880
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 0.12096,
|
| 1342 |
+
"grad_norm": 1.1372530460357666,
|
| 1343 |
+
"learning_rate": 8.060593129933861e-06,
|
| 1344 |
+
"loss": 0.6264,
|
| 1345 |
+
"step": 1890
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 0.1216,
|
| 1349 |
+
"grad_norm": 1.0789086818695068,
|
| 1350 |
+
"learning_rate": 8.103264348197143e-06,
|
| 1351 |
+
"loss": 0.6396,
|
| 1352 |
+
"step": 1900
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 0.12224,
|
| 1356 |
+
"grad_norm": 1.3709907531738281,
|
| 1357 |
+
"learning_rate": 8.145935566460424e-06,
|
| 1358 |
+
"loss": 0.638,
|
| 1359 |
+
"step": 1910
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 0.12288,
|
| 1363 |
+
"grad_norm": 1.5928348302841187,
|
| 1364 |
+
"learning_rate": 8.188606784723706e-06,
|
| 1365 |
+
"loss": 0.6495,
|
| 1366 |
+
"step": 1920
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 0.12352,
|
| 1370 |
+
"grad_norm": 1.0440688133239746,
|
| 1371 |
+
"learning_rate": 8.231278002986986e-06,
|
| 1372 |
+
"loss": 0.6236,
|
| 1373 |
+
"step": 1930
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 0.12416,
|
| 1377 |
+
"grad_norm": 1.2079739570617676,
|
| 1378 |
+
"learning_rate": 8.273949221250267e-06,
|
| 1379 |
+
"loss": 0.6489,
|
| 1380 |
+
"step": 1940
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 0.1248,
|
| 1384 |
+
"grad_norm": 1.161354422569275,
|
| 1385 |
+
"learning_rate": 8.316620439513549e-06,
|
| 1386 |
+
"loss": 0.6412,
|
| 1387 |
+
"step": 1950
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 0.12544,
|
| 1391 |
+
"grad_norm": 1.7108139991760254,
|
| 1392 |
+
"learning_rate": 8.35929165777683e-06,
|
| 1393 |
+
"loss": 0.6383,
|
| 1394 |
+
"step": 1960
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 0.12608,
|
| 1398 |
+
"grad_norm": 1.4153392314910889,
|
| 1399 |
+
"learning_rate": 8.401962876040112e-06,
|
| 1400 |
+
"loss": 0.6482,
|
| 1401 |
+
"step": 1970
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 0.12672,
|
| 1405 |
+
"grad_norm": 1.6085275411605835,
|
| 1406 |
+
"learning_rate": 8.444634094303393e-06,
|
| 1407 |
+
"loss": 0.6488,
|
| 1408 |
+
"step": 1980
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 0.12736,
|
| 1412 |
+
"grad_norm": 1.2328271865844727,
|
| 1413 |
+
"learning_rate": 8.487305312566675e-06,
|
| 1414 |
+
"loss": 0.6466,
|
| 1415 |
+
"step": 1990
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 0.128,
|
| 1419 |
+
"grad_norm": 0.9865418672561646,
|
| 1420 |
+
"learning_rate": 8.529976530829955e-06,
|
| 1421 |
+
"loss": 0.6468,
|
| 1422 |
+
"step": 2000
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 0.128,
|
| 1426 |
+
"eval_accuracy": 0.7972,
|
| 1427 |
+
"eval_loss": 0.578827440738678,
|
| 1428 |
+
"eval_macro_f1": 0.797186028145479,
|
| 1429 |
+
"eval_precision": 0.7981394308396883,
|
| 1430 |
+
"eval_recall": 0.7985143807471076,
|
| 1431 |
+
"eval_runtime": 813.4399,
|
| 1432 |
+
"eval_samples_per_second": 122.935,
|
| 1433 |
+
"eval_steps_per_second": 1.921,
|
| 1434 |
+
"step": 2000
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 0.12864,
|
| 1438 |
+
"grad_norm": 1.0044854879379272,
|
| 1439 |
+
"learning_rate": 8.572647749093236e-06,
|
| 1440 |
+
"loss": 0.6464,
|
| 1441 |
+
"step": 2010
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 0.12928,
|
| 1445 |
+
"grad_norm": 1.3742347955703735,
|
| 1446 |
+
"learning_rate": 8.615318967356518e-06,
|
| 1447 |
+
"loss": 0.6373,
|
| 1448 |
+
"step": 2020
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 0.12992,
|
| 1452 |
+
"grad_norm": 1.3543511629104614,
|
| 1453 |
+
"learning_rate": 8.6579901856198e-06,
|
| 1454 |
+
"loss": 0.6344,
|
| 1455 |
+
"step": 2030
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 0.13056,
|
| 1459 |
+
"grad_norm": 1.865698218345642,
|
| 1460 |
+
"learning_rate": 8.700661403883081e-06,
|
| 1461 |
+
"loss": 0.6426,
|
| 1462 |
+
"step": 2040
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 0.1312,
|
| 1466 |
+
"grad_norm": 1.3193360567092896,
|
| 1467 |
+
"learning_rate": 8.743332622146363e-06,
|
| 1468 |
+
"loss": 0.6351,
|
| 1469 |
+
"step": 2050
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 0.13184,
|
| 1473 |
+
"grad_norm": 1.5337750911712646,
|
| 1474 |
+
"learning_rate": 8.786003840409644e-06,
|
| 1475 |
+
"loss": 0.6374,
|
| 1476 |
+
"step": 2060
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 0.13248,
|
| 1480 |
+
"grad_norm": 1.7162808179855347,
|
| 1481 |
+
"learning_rate": 8.828675058672926e-06,
|
| 1482 |
+
"loss": 0.6445,
|
| 1483 |
+
"step": 2070
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 0.13312,
|
| 1487 |
+
"grad_norm": 1.9505610466003418,
|
| 1488 |
+
"learning_rate": 8.871346276936208e-06,
|
| 1489 |
+
"loss": 0.6223,
|
| 1490 |
+
"step": 2080
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 0.13376,
|
| 1494 |
+
"grad_norm": 2.593158721923828,
|
| 1495 |
+
"learning_rate": 8.914017495199489e-06,
|
| 1496 |
+
"loss": 0.6272,
|
| 1497 |
+
"step": 2090
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 0.1344,
|
| 1501 |
+
"grad_norm": 0.9756277203559875,
|
| 1502 |
+
"learning_rate": 8.95668871346277e-06,
|
| 1503 |
+
"loss": 0.6383,
|
| 1504 |
+
"step": 2100
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 0.13504,
|
| 1508 |
+
"grad_norm": 1.3016871213912964,
|
| 1509 |
+
"learning_rate": 8.999359931726052e-06,
|
| 1510 |
+
"loss": 0.6208,
|
| 1511 |
+
"step": 2110
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 0.13568,
|
| 1515 |
+
"grad_norm": 1.475502610206604,
|
| 1516 |
+
"learning_rate": 9.042031149989334e-06,
|
| 1517 |
+
"loss": 0.6354,
|
| 1518 |
+
"step": 2120
|
| 1519 |
+
},
|
| 1520 |
+
{
|
| 1521 |
+
"epoch": 0.13632,
|
| 1522 |
+
"grad_norm": 1.6524313688278198,
|
| 1523 |
+
"learning_rate": 9.084702368252615e-06,
|
| 1524 |
+
"loss": 0.6515,
|
| 1525 |
+
"step": 2130
|
| 1526 |
+
},
|
| 1527 |
+
{
|
| 1528 |
+
"epoch": 0.13696,
|
| 1529 |
+
"grad_norm": 1.5008305311203003,
|
| 1530 |
+
"learning_rate": 9.127373586515895e-06,
|
| 1531 |
+
"loss": 0.6468,
|
| 1532 |
+
"step": 2140
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 0.1376,
|
| 1536 |
+
"grad_norm": 1.2291077375411987,
|
| 1537 |
+
"learning_rate": 9.170044804779177e-06,
|
| 1538 |
+
"loss": 0.6271,
|
| 1539 |
+
"step": 2150
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 0.13824,
|
| 1543 |
+
"grad_norm": 1.5787503719329834,
|
| 1544 |
+
"learning_rate": 9.212716023042458e-06,
|
| 1545 |
+
"loss": 0.6495,
|
| 1546 |
+
"step": 2160
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 0.13888,
|
| 1550 |
+
"grad_norm": 1.6643162965774536,
|
| 1551 |
+
"learning_rate": 9.25538724130574e-06,
|
| 1552 |
+
"loss": 0.6417,
|
| 1553 |
+
"step": 2170
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 0.13952,
|
| 1557 |
+
"grad_norm": 1.5859993696212769,
|
| 1558 |
+
"learning_rate": 9.298058459569022e-06,
|
| 1559 |
+
"loss": 0.6362,
|
| 1560 |
+
"step": 2180
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 0.14016,
|
| 1564 |
+
"grad_norm": 1.2435864210128784,
|
| 1565 |
+
"learning_rate": 9.340729677832303e-06,
|
| 1566 |
+
"loss": 0.6337,
|
| 1567 |
+
"step": 2190
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"epoch": 0.1408,
|
| 1571 |
+
"grad_norm": 1.1807382106781006,
|
| 1572 |
+
"learning_rate": 9.383400896095585e-06,
|
| 1573 |
+
"loss": 0.6344,
|
| 1574 |
+
"step": 2200
|
| 1575 |
+
}
|
| 1576 |
+
],
|
| 1577 |
+
"logging_steps": 10,
|
| 1578 |
+
"max_steps": 46875,
|
| 1579 |
+
"num_input_tokens_seen": 0,
|
| 1580 |
+
"num_train_epochs": 3,
|
| 1581 |
+
"save_steps": 200,
|
| 1582 |
+
"stateful_callbacks": {
|
| 1583 |
+
"EarlyStoppingCallback": {
|
| 1584 |
+
"args": {
|
| 1585 |
+
"early_stopping_patience": 3,
|
| 1586 |
+
"early_stopping_threshold": 0.0
|
| 1587 |
+
},
|
| 1588 |
+
"attributes": {
|
| 1589 |
+
"early_stopping_patience_counter": 0
|
| 1590 |
+
}
|
| 1591 |
+
},
|
| 1592 |
+
"TrainerControl": {
|
| 1593 |
+
"args": {
|
| 1594 |
+
"should_epoch_stop": false,
|
| 1595 |
+
"should_evaluate": false,
|
| 1596 |
+
"should_log": false,
|
| 1597 |
+
"should_save": true,
|
| 1598 |
+
"should_training_stop": false
|
| 1599 |
+
},
|
| 1600 |
+
"attributes": {}
|
| 1601 |
+
}
|
| 1602 |
+
},
|
| 1603 |
+
"total_flos": 1.851581563220352e+16,
|
| 1604 |
+
"train_batch_size": 32,
|
| 1605 |
+
"trial_name": null,
|
| 1606 |
+
"trial_params": null
|
| 1607 |
+
}
|