Upload 41 files
Browse files- config.json +26 -0
- merges.txt +0 -0
- rng_state_0.pth +3 -0
- rng_state_1.pth +3 -0
- rng_state_10.pth +3 -0
- rng_state_11.pth +3 -0
- rng_state_12.pth +3 -0
- rng_state_13.pth +3 -0
- rng_state_14.pth +3 -0
- rng_state_15.pth +3 -0
- rng_state_16.pth +3 -0
- rng_state_17.pth +3 -0
- rng_state_18.pth +3 -0
- rng_state_19.pth +3 -0
- rng_state_2.pth +3 -0
- rng_state_20.pth +3 -0
- rng_state_21.pth +3 -0
- rng_state_22.pth +3 -0
- rng_state_23.pth +3 -0
- rng_state_24.pth +3 -0
- rng_state_25.pth +3 -0
- rng_state_26.pth +3 -0
- rng_state_27.pth +3 -0
- rng_state_28.pth +3 -0
- rng_state_29.pth +3 -0
- rng_state_3.pth +3 -0
- rng_state_30.pth +3 -0
- rng_state_31.pth +3 -0
- rng_state_4.pth +3 -0
- rng_state_5.pth +3 -0
- rng_state_6.pth +3 -0
- rng_state_7.pth +3 -0
- rng_state_8.pth +3 -0
- rng_state_9.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +51 -0
- tokenizer_config.json +66 -0
- trainer_state.json +1198 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"CamembertForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-12,
|
| 15 |
+
"max_position_embeddings": 514,
|
| 16 |
+
"model_type": "camembert",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 1,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.22.2",
|
| 23 |
+
"type_vocab_size": 1,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 32000
|
| 26 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_10.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_11.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_12.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_13.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_14.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_15.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_16.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_17.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_18.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_19.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_20.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_21.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_22.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_23.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_24.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_25.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_26.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_27.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_28.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_29.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_30.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_31.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_8.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
rng_state_9.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be0b287b14419a0d5070e8035acd0874e276253435a4621b895a94b5c1e43d8f
|
| 3 |
+
size 14503
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3415138cb73a5883c98f0bf2c8f14e8096a4e812f12f4501f88e16af49d22e3e
|
| 3 |
+
size 559
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:97870f9ebbb5b7309e3077b264e82949212c1b705eb2a704a9488ea1cd27c4ad
|
| 3 |
+
size 623
|
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 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,66 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": {
|
| 4 |
+
"__type": "AddedToken",
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false
|
| 10 |
+
},
|
| 11 |
+
"clean_up_tokenization_spaces": true,
|
| 12 |
+
"cls_token": {
|
| 13 |
+
"__type": "AddedToken",
|
| 14 |
+
"content": "<s>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": true,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false
|
| 19 |
+
},
|
| 20 |
+
"eos_token": {
|
| 21 |
+
"__type": "AddedToken",
|
| 22 |
+
"content": "</s>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": true,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false
|
| 27 |
+
},
|
| 28 |
+
"errors": "replace",
|
| 29 |
+
"mask_token": {
|
| 30 |
+
"__type": "AddedToken",
|
| 31 |
+
"content": "<mask>",
|
| 32 |
+
"lstrip": true,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 38 |
+
"name_or_path": "./tokenizers/BpeTokenizer_Wikipedia_CHARS_lowercased_fixed_utf8-roberta/",
|
| 39 |
+
"pad_token": {
|
| 40 |
+
"__type": "AddedToken",
|
| 41 |
+
"content": "<pad>",
|
| 42 |
+
"lstrip": false,
|
| 43 |
+
"normalized": true,
|
| 44 |
+
"rstrip": false,
|
| 45 |
+
"single_word": false
|
| 46 |
+
},
|
| 47 |
+
"sep_token": {
|
| 48 |
+
"__type": "AddedToken",
|
| 49 |
+
"content": "</s>",
|
| 50 |
+
"lstrip": false,
|
| 51 |
+
"normalized": true,
|
| 52 |
+
"rstrip": false,
|
| 53 |
+
"single_word": false
|
| 54 |
+
},
|
| 55 |
+
"special_tokens_map_file": "./tokenizers/BpeTokenizer_Wikipedia_CHARS_lowercased_fixed_utf8-roberta/special_tokens_map.json",
|
| 56 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 57 |
+
"unk_token": {
|
| 58 |
+
"__type": "AddedToken",
|
| 59 |
+
"content": "<unk>",
|
| 60 |
+
"lstrip": false,
|
| 61 |
+
"normalized": true,
|
| 62 |
+
"rstrip": false,
|
| 63 |
+
"single_word": false
|
| 64 |
+
},
|
| 65 |
+
"use_fast": true
|
| 66 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1198 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 26.0,
|
| 5 |
+
"global_step": 98228,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 5e-09,
|
| 13 |
+
"loss": 10.4629,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.13,
|
| 18 |
+
"learning_rate": 2.5e-06,
|
| 19 |
+
"loss": 9.562,
|
| 20 |
+
"step": 500
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.26,
|
| 24 |
+
"learning_rate": 5e-06,
|
| 25 |
+
"loss": 7.9982,
|
| 26 |
+
"step": 1000
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.4,
|
| 30 |
+
"learning_rate": 7.5e-06,
|
| 31 |
+
"loss": 6.9798,
|
| 32 |
+
"step": 1500
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.53,
|
| 36 |
+
"learning_rate": 1e-05,
|
| 37 |
+
"loss": 6.6752,
|
| 38 |
+
"step": 2000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.66,
|
| 42 |
+
"learning_rate": 1.25e-05,
|
| 43 |
+
"loss": 6.5045,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.79,
|
| 48 |
+
"learning_rate": 1.5e-05,
|
| 49 |
+
"loss": 6.3892,
|
| 50 |
+
"step": 3000
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.93,
|
| 54 |
+
"learning_rate": 1.75e-05,
|
| 55 |
+
"loss": 6.2943,
|
| 56 |
+
"step": 3500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.06,
|
| 60 |
+
"learning_rate": 2e-05,
|
| 61 |
+
"loss": 6.2221,
|
| 62 |
+
"step": 4000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.19,
|
| 66 |
+
"learning_rate": 2.25e-05,
|
| 67 |
+
"loss": 6.154,
|
| 68 |
+
"step": 4500
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.32,
|
| 72 |
+
"learning_rate": 2.5e-05,
|
| 73 |
+
"loss": 6.0982,
|
| 74 |
+
"step": 5000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.46,
|
| 78 |
+
"learning_rate": 2.7500000000000004e-05,
|
| 79 |
+
"loss": 6.0464,
|
| 80 |
+
"step": 5500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.59,
|
| 84 |
+
"learning_rate": 3e-05,
|
| 85 |
+
"loss": 6.0017,
|
| 86 |
+
"step": 6000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.72,
|
| 90 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 91 |
+
"loss": 5.9683,
|
| 92 |
+
"step": 6500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.85,
|
| 96 |
+
"learning_rate": 3.5e-05,
|
| 97 |
+
"loss": 5.936,
|
| 98 |
+
"step": 7000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.99,
|
| 102 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 103 |
+
"loss": 5.9005,
|
| 104 |
+
"step": 7500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.12,
|
| 108 |
+
"learning_rate": 4e-05,
|
| 109 |
+
"loss": 5.8763,
|
| 110 |
+
"step": 8000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 2.25,
|
| 114 |
+
"learning_rate": 4.2495e-05,
|
| 115 |
+
"loss": 5.8522,
|
| 116 |
+
"step": 8500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.38,
|
| 120 |
+
"learning_rate": 4.4995000000000005e-05,
|
| 121 |
+
"loss": 5.8356,
|
| 122 |
+
"step": 9000
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 2.51,
|
| 126 |
+
"learning_rate": 4.7495e-05,
|
| 127 |
+
"loss": 5.8125,
|
| 128 |
+
"step": 9500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.65,
|
| 132 |
+
"learning_rate": 4.9995000000000005e-05,
|
| 133 |
+
"loss": 5.7984,
|
| 134 |
+
"step": 10000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.78,
|
| 138 |
+
"learning_rate": 4.998341326938449e-05,
|
| 139 |
+
"loss": 5.7788,
|
| 140 |
+
"step": 10500
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 2.91,
|
| 144 |
+
"learning_rate": 4.996675992539302e-05,
|
| 145 |
+
"loss": 5.7609,
|
| 146 |
+
"step": 11000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 3.04,
|
| 150 |
+
"learning_rate": 4.995010658140154e-05,
|
| 151 |
+
"loss": 5.7509,
|
| 152 |
+
"step": 11500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 3.18,
|
| 156 |
+
"learning_rate": 4.993345323741007e-05,
|
| 157 |
+
"loss": 5.7398,
|
| 158 |
+
"step": 12000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 3.31,
|
| 162 |
+
"learning_rate": 4.991683320010658e-05,
|
| 163 |
+
"loss": 5.7264,
|
| 164 |
+
"step": 12500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.44,
|
| 168 |
+
"learning_rate": 4.990017985611511e-05,
|
| 169 |
+
"loss": 5.7196,
|
| 170 |
+
"step": 13000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.57,
|
| 174 |
+
"learning_rate": 4.988352651212364e-05,
|
| 175 |
+
"loss": 5.7094,
|
| 176 |
+
"step": 13500
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 3.71,
|
| 180 |
+
"learning_rate": 4.986687316813216e-05,
|
| 181 |
+
"loss": 5.6992,
|
| 182 |
+
"step": 14000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 3.84,
|
| 186 |
+
"learning_rate": 4.9850253130828675e-05,
|
| 187 |
+
"loss": 5.6917,
|
| 188 |
+
"step": 14500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 3.97,
|
| 192 |
+
"learning_rate": 4.9833599786837196e-05,
|
| 193 |
+
"loss": 5.6858,
|
| 194 |
+
"step": 15000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 4.1,
|
| 198 |
+
"learning_rate": 4.9816946442845725e-05,
|
| 199 |
+
"loss": 5.6734,
|
| 200 |
+
"step": 15500
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 4.24,
|
| 204 |
+
"learning_rate": 4.980029309885425e-05,
|
| 205 |
+
"loss": 5.6687,
|
| 206 |
+
"step": 16000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 4.37,
|
| 210 |
+
"learning_rate": 4.978367306155076e-05,
|
| 211 |
+
"loss": 5.6662,
|
| 212 |
+
"step": 16500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 4.5,
|
| 216 |
+
"learning_rate": 4.976701971755929e-05,
|
| 217 |
+
"loss": 5.6598,
|
| 218 |
+
"step": 17000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 4.63,
|
| 222 |
+
"learning_rate": 4.975036637356781e-05,
|
| 223 |
+
"loss": 5.655,
|
| 224 |
+
"step": 17500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 4.76,
|
| 228 |
+
"learning_rate": 4.973374633626433e-05,
|
| 229 |
+
"loss": 5.6471,
|
| 230 |
+
"step": 18000
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 4.9,
|
| 234 |
+
"learning_rate": 4.971709299227285e-05,
|
| 235 |
+
"loss": 5.6424,
|
| 236 |
+
"step": 18500
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 5.03,
|
| 240 |
+
"learning_rate": 4.970043964828138e-05,
|
| 241 |
+
"loss": 5.6371,
|
| 242 |
+
"step": 19000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 5.16,
|
| 246 |
+
"learning_rate": 4.96837863042899e-05,
|
| 247 |
+
"loss": 5.6289,
|
| 248 |
+
"step": 19500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 5.29,
|
| 252 |
+
"learning_rate": 4.966713296029843e-05,
|
| 253 |
+
"loss": 5.6248,
|
| 254 |
+
"step": 20000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 5.43,
|
| 258 |
+
"learning_rate": 4.965047961630696e-05,
|
| 259 |
+
"loss": 5.622,
|
| 260 |
+
"step": 20500
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 5.56,
|
| 264 |
+
"learning_rate": 4.963382627231548e-05,
|
| 265 |
+
"loss": 5.6202,
|
| 266 |
+
"step": 21000
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 5.69,
|
| 270 |
+
"learning_rate": 4.9617206235011995e-05,
|
| 271 |
+
"loss": 5.6134,
|
| 272 |
+
"step": 21500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 5.82,
|
| 276 |
+
"learning_rate": 4.9600552891020516e-05,
|
| 277 |
+
"loss": 5.6098,
|
| 278 |
+
"step": 22000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 5.96,
|
| 282 |
+
"learning_rate": 4.9583899547029045e-05,
|
| 283 |
+
"loss": 5.6084,
|
| 284 |
+
"step": 22500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 6.09,
|
| 288 |
+
"learning_rate": 4.956724620303757e-05,
|
| 289 |
+
"loss": 5.6018,
|
| 290 |
+
"step": 23000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 6.22,
|
| 294 |
+
"learning_rate": 4.9550592859046095e-05,
|
| 295 |
+
"loss": 5.5955,
|
| 296 |
+
"step": 23500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 6.35,
|
| 300 |
+
"learning_rate": 4.953397282174261e-05,
|
| 301 |
+
"loss": 5.5957,
|
| 302 |
+
"step": 24000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 6.48,
|
| 306 |
+
"learning_rate": 4.951731947775113e-05,
|
| 307 |
+
"loss": 5.5875,
|
| 308 |
+
"step": 24500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 6.62,
|
| 312 |
+
"learning_rate": 4.950066613375966e-05,
|
| 313 |
+
"loss": 5.5857,
|
| 314 |
+
"step": 25000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 6.75,
|
| 318 |
+
"learning_rate": 4.948401278976819e-05,
|
| 319 |
+
"loss": 5.5856,
|
| 320 |
+
"step": 25500
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 6.88,
|
| 324 |
+
"learning_rate": 4.94673927524647e-05,
|
| 325 |
+
"loss": 5.5855,
|
| 326 |
+
"step": 26000
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 7.01,
|
| 330 |
+
"learning_rate": 4.945073940847322e-05,
|
| 331 |
+
"loss": 5.579,
|
| 332 |
+
"step": 26500
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 7.15,
|
| 336 |
+
"learning_rate": 4.943408606448175e-05,
|
| 337 |
+
"loss": 5.5772,
|
| 338 |
+
"step": 27000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 7.28,
|
| 342 |
+
"learning_rate": 4.941743272049028e-05,
|
| 343 |
+
"loss": 5.5725,
|
| 344 |
+
"step": 27500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 7.41,
|
| 348 |
+
"learning_rate": 4.94007793764988e-05,
|
| 349 |
+
"loss": 5.57,
|
| 350 |
+
"step": 28000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 7.54,
|
| 354 |
+
"learning_rate": 4.9384159339195315e-05,
|
| 355 |
+
"loss": 5.5676,
|
| 356 |
+
"step": 28500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 7.68,
|
| 360 |
+
"learning_rate": 4.9367505995203836e-05,
|
| 361 |
+
"loss": 5.5665,
|
| 362 |
+
"step": 29000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 7.81,
|
| 366 |
+
"learning_rate": 4.9350852651212365e-05,
|
| 367 |
+
"loss": 5.5615,
|
| 368 |
+
"step": 29500
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 7.94,
|
| 372 |
+
"learning_rate": 4.933419930722089e-05,
|
| 373 |
+
"loss": 5.5169,
|
| 374 |
+
"step": 30000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 8.07,
|
| 378 |
+
"learning_rate": 4.9317545963229415e-05,
|
| 379 |
+
"loss": 5.2484,
|
| 380 |
+
"step": 30500
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 8.21,
|
| 384 |
+
"learning_rate": 4.930089261923794e-05,
|
| 385 |
+
"loss": 5.043,
|
| 386 |
+
"step": 31000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 8.34,
|
| 390 |
+
"learning_rate": 4.928423927524647e-05,
|
| 391 |
+
"loss": 4.8676,
|
| 392 |
+
"step": 31500
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 8.47,
|
| 396 |
+
"learning_rate": 4.9267585931255e-05,
|
| 397 |
+
"loss": 4.6987,
|
| 398 |
+
"step": 32000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 8.6,
|
| 402 |
+
"learning_rate": 4.925096589395151e-05,
|
| 403 |
+
"loss": 4.5359,
|
| 404 |
+
"step": 32500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 8.73,
|
| 408 |
+
"learning_rate": 4.923434585664802e-05,
|
| 409 |
+
"loss": 4.3603,
|
| 410 |
+
"step": 33000
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 8.87,
|
| 414 |
+
"learning_rate": 4.921769251265655e-05,
|
| 415 |
+
"loss": 4.1923,
|
| 416 |
+
"step": 33500
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 9.0,
|
| 420 |
+
"learning_rate": 4.920103916866507e-05,
|
| 421 |
+
"loss": 4.0239,
|
| 422 |
+
"step": 34000
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 9.13,
|
| 426 |
+
"learning_rate": 4.91843858246736e-05,
|
| 427 |
+
"loss": 3.7255,
|
| 428 |
+
"step": 34500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 9.26,
|
| 432 |
+
"learning_rate": 4.916773248068212e-05,
|
| 433 |
+
"loss": 3.2457,
|
| 434 |
+
"step": 35000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 9.4,
|
| 438 |
+
"learning_rate": 4.9151112443378634e-05,
|
| 439 |
+
"loss": 2.5753,
|
| 440 |
+
"step": 35500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 9.53,
|
| 444 |
+
"learning_rate": 4.9134459099387156e-05,
|
| 445 |
+
"loss": 2.2821,
|
| 446 |
+
"step": 36000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 9.66,
|
| 450 |
+
"learning_rate": 4.9117805755395685e-05,
|
| 451 |
+
"loss": 2.1267,
|
| 452 |
+
"step": 36500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 9.79,
|
| 456 |
+
"learning_rate": 4.910115241140421e-05,
|
| 457 |
+
"loss": 2.0179,
|
| 458 |
+
"step": 37000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 9.93,
|
| 462 |
+
"learning_rate": 4.908453237410072e-05,
|
| 463 |
+
"loss": 1.9368,
|
| 464 |
+
"step": 37500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 10.06,
|
| 468 |
+
"learning_rate": 4.906787903010925e-05,
|
| 469 |
+
"loss": 1.8691,
|
| 470 |
+
"step": 38000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 10.19,
|
| 474 |
+
"learning_rate": 4.905125899280576e-05,
|
| 475 |
+
"loss": 1.816,
|
| 476 |
+
"step": 38500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 10.32,
|
| 480 |
+
"learning_rate": 4.903460564881429e-05,
|
| 481 |
+
"loss": 1.7673,
|
| 482 |
+
"step": 39000
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 10.46,
|
| 486 |
+
"learning_rate": 4.901795230482281e-05,
|
| 487 |
+
"loss": 1.7288,
|
| 488 |
+
"step": 39500
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 10.59,
|
| 492 |
+
"learning_rate": 4.900129896083134e-05,
|
| 493 |
+
"loss": 1.6907,
|
| 494 |
+
"step": 40000
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 10.72,
|
| 498 |
+
"learning_rate": 4.898464561683987e-05,
|
| 499 |
+
"loss": 1.6572,
|
| 500 |
+
"step": 40500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 10.85,
|
| 504 |
+
"learning_rate": 4.896799227284839e-05,
|
| 505 |
+
"loss": 1.6261,
|
| 506 |
+
"step": 41000
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 10.98,
|
| 510 |
+
"learning_rate": 4.895133892885692e-05,
|
| 511 |
+
"loss": 1.6001,
|
| 512 |
+
"step": 41500
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 11.12,
|
| 516 |
+
"learning_rate": 4.893468558486544e-05,
|
| 517 |
+
"loss": 1.5719,
|
| 518 |
+
"step": 42000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 11.25,
|
| 522 |
+
"learning_rate": 4.891803224087397e-05,
|
| 523 |
+
"loss": 1.5495,
|
| 524 |
+
"step": 42500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 11.38,
|
| 528 |
+
"learning_rate": 4.8901412203570476e-05,
|
| 529 |
+
"loss": 1.5277,
|
| 530 |
+
"step": 43000
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 11.51,
|
| 534 |
+
"learning_rate": 4.8884758859579004e-05,
|
| 535 |
+
"loss": 1.5069,
|
| 536 |
+
"step": 43500
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 11.65,
|
| 540 |
+
"learning_rate": 4.886810551558753e-05,
|
| 541 |
+
"loss": 1.4871,
|
| 542 |
+
"step": 44000
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 11.78,
|
| 546 |
+
"learning_rate": 4.8851452171596055e-05,
|
| 547 |
+
"loss": 1.4707,
|
| 548 |
+
"step": 44500
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 11.91,
|
| 552 |
+
"learning_rate": 4.883479882760458e-05,
|
| 553 |
+
"loss": 1.4522,
|
| 554 |
+
"step": 45000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 12.04,
|
| 558 |
+
"learning_rate": 4.881817879030109e-05,
|
| 559 |
+
"loss": 1.4362,
|
| 560 |
+
"step": 45500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 12.18,
|
| 564 |
+
"learning_rate": 4.880152544630962e-05,
|
| 565 |
+
"loss": 1.4219,
|
| 566 |
+
"step": 46000
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 12.31,
|
| 570 |
+
"learning_rate": 4.878487210231815e-05,
|
| 571 |
+
"loss": 1.4066,
|
| 572 |
+
"step": 46500
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 12.44,
|
| 576 |
+
"learning_rate": 4.876821875832667e-05,
|
| 577 |
+
"loss": 1.3935,
|
| 578 |
+
"step": 47000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 12.57,
|
| 582 |
+
"learning_rate": 4.875159872102319e-05,
|
| 583 |
+
"loss": 1.3809,
|
| 584 |
+
"step": 47500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 12.71,
|
| 588 |
+
"learning_rate": 4.873494537703171e-05,
|
| 589 |
+
"loss": 1.3663,
|
| 590 |
+
"step": 48000
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 12.84,
|
| 594 |
+
"learning_rate": 4.871829203304024e-05,
|
| 595 |
+
"loss": 1.3571,
|
| 596 |
+
"step": 48500
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 12.97,
|
| 600 |
+
"learning_rate": 4.8701671995736746e-05,
|
| 601 |
+
"loss": 1.3454,
|
| 602 |
+
"step": 49000
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 13.1,
|
| 606 |
+
"learning_rate": 4.8685018651745274e-05,
|
| 607 |
+
"loss": 1.3336,
|
| 608 |
+
"step": 49500
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 13.23,
|
| 612 |
+
"learning_rate": 4.8668365307753796e-05,
|
| 613 |
+
"loss": 1.3236,
|
| 614 |
+
"step": 50000
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 13.37,
|
| 618 |
+
"learning_rate": 4.8651711963762324e-05,
|
| 619 |
+
"loss": 1.3131,
|
| 620 |
+
"step": 50500
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 13.5,
|
| 624 |
+
"learning_rate": 4.863505861977085e-05,
|
| 625 |
+
"loss": 1.3029,
|
| 626 |
+
"step": 51000
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 13.63,
|
| 630 |
+
"learning_rate": 4.8618405275779374e-05,
|
| 631 |
+
"loss": 1.2961,
|
| 632 |
+
"step": 51500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 13.76,
|
| 636 |
+
"learning_rate": 4.86017519317879e-05,
|
| 637 |
+
"loss": 1.2889,
|
| 638 |
+
"step": 52000
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 13.9,
|
| 642 |
+
"learning_rate": 4.858513189448441e-05,
|
| 643 |
+
"loss": 1.2814,
|
| 644 |
+
"step": 52500
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 14.03,
|
| 648 |
+
"learning_rate": 4.856847855049294e-05,
|
| 649 |
+
"loss": 1.2704,
|
| 650 |
+
"step": 53000
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 14.16,
|
| 654 |
+
"learning_rate": 4.855182520650147e-05,
|
| 655 |
+
"loss": 1.2605,
|
| 656 |
+
"step": 53500
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 14.29,
|
| 660 |
+
"learning_rate": 4.853517186250999e-05,
|
| 661 |
+
"loss": 1.2545,
|
| 662 |
+
"step": 54000
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 14.43,
|
| 666 |
+
"learning_rate": 4.851851851851852e-05,
|
| 667 |
+
"loss": 1.2487,
|
| 668 |
+
"step": 54500
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 14.56,
|
| 672 |
+
"learning_rate": 4.850186517452705e-05,
|
| 673 |
+
"loss": 1.2381,
|
| 674 |
+
"step": 55000
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 14.69,
|
| 678 |
+
"learning_rate": 4.8485211830535576e-05,
|
| 679 |
+
"loss": 1.2326,
|
| 680 |
+
"step": 55500
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 14.82,
|
| 684 |
+
"learning_rate": 4.8468558486544104e-05,
|
| 685 |
+
"loss": 1.224,
|
| 686 |
+
"step": 56000
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 14.96,
|
| 690 |
+
"learning_rate": 4.845193844924061e-05,
|
| 691 |
+
"loss": 1.2213,
|
| 692 |
+
"step": 56500
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 15.09,
|
| 696 |
+
"learning_rate": 4.8435285105249135e-05,
|
| 697 |
+
"loss": 1.2126,
|
| 698 |
+
"step": 57000
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 15.22,
|
| 702 |
+
"learning_rate": 4.841863176125766e-05,
|
| 703 |
+
"loss": 1.2047,
|
| 704 |
+
"step": 57500
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 15.35,
|
| 708 |
+
"learning_rate": 4.840197841726619e-05,
|
| 709 |
+
"loss": 1.1999,
|
| 710 |
+
"step": 58000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 15.48,
|
| 714 |
+
"learning_rate": 4.8385358379962694e-05,
|
| 715 |
+
"loss": 1.1967,
|
| 716 |
+
"step": 58500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 15.62,
|
| 720 |
+
"learning_rate": 4.836870503597122e-05,
|
| 721 |
+
"loss": 1.1901,
|
| 722 |
+
"step": 59000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 15.75,
|
| 726 |
+
"learning_rate": 4.835205169197975e-05,
|
| 727 |
+
"loss": 1.1843,
|
| 728 |
+
"step": 59500
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 15.88,
|
| 732 |
+
"learning_rate": 4.833539834798828e-05,
|
| 733 |
+
"loss": 1.1792,
|
| 734 |
+
"step": 60000
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 16.01,
|
| 738 |
+
"learning_rate": 4.831877831068479e-05,
|
| 739 |
+
"loss": 1.1752,
|
| 740 |
+
"step": 60500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 16.15,
|
| 744 |
+
"learning_rate": 4.83021582733813e-05,
|
| 745 |
+
"loss": 1.1657,
|
| 746 |
+
"step": 61000
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 16.28,
|
| 750 |
+
"learning_rate": 4.8285504929389827e-05,
|
| 751 |
+
"loss": 1.1628,
|
| 752 |
+
"step": 61500
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 16.41,
|
| 756 |
+
"learning_rate": 4.826885158539835e-05,
|
| 757 |
+
"loss": 1.1586,
|
| 758 |
+
"step": 62000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 16.54,
|
| 762 |
+
"learning_rate": 4.825219824140688e-05,
|
| 763 |
+
"loss": 1.1547,
|
| 764 |
+
"step": 62500
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 16.68,
|
| 768 |
+
"learning_rate": 4.8235578204103386e-05,
|
| 769 |
+
"loss": 1.1486,
|
| 770 |
+
"step": 63000
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 16.81,
|
| 774 |
+
"learning_rate": 4.8218924860111914e-05,
|
| 775 |
+
"loss": 1.1439,
|
| 776 |
+
"step": 63500
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 16.94,
|
| 780 |
+
"learning_rate": 4.820227151612044e-05,
|
| 781 |
+
"loss": 1.1409,
|
| 782 |
+
"step": 64000
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 17.07,
|
| 786 |
+
"learning_rate": 4.8185618172128964e-05,
|
| 787 |
+
"loss": 1.1349,
|
| 788 |
+
"step": 64500
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 17.2,
|
| 792 |
+
"learning_rate": 4.816896482813749e-05,
|
| 793 |
+
"loss": 1.1306,
|
| 794 |
+
"step": 65000
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 17.34,
|
| 798 |
+
"learning_rate": 4.8152311484146014e-05,
|
| 799 |
+
"loss": 1.1287,
|
| 800 |
+
"step": 65500
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 17.47,
|
| 804 |
+
"learning_rate": 4.813569144684253e-05,
|
| 805 |
+
"loss": 1.1224,
|
| 806 |
+
"step": 66000
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 17.6,
|
| 810 |
+
"learning_rate": 4.811903810285105e-05,
|
| 811 |
+
"loss": 1.1196,
|
| 812 |
+
"step": 66500
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 17.73,
|
| 816 |
+
"learning_rate": 4.810238475885958e-05,
|
| 817 |
+
"loss": 1.1162,
|
| 818 |
+
"step": 67000
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 17.87,
|
| 822 |
+
"learning_rate": 4.808576472155609e-05,
|
| 823 |
+
"loss": 1.1141,
|
| 824 |
+
"step": 67500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 18.0,
|
| 828 |
+
"learning_rate": 4.806911137756462e-05,
|
| 829 |
+
"loss": 1.1098,
|
| 830 |
+
"step": 68000
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 18.13,
|
| 834 |
+
"learning_rate": 4.8052458033573147e-05,
|
| 835 |
+
"loss": 1.1043,
|
| 836 |
+
"step": 68500
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 18.26,
|
| 840 |
+
"learning_rate": 4.803580468958167e-05,
|
| 841 |
+
"loss": 1.1008,
|
| 842 |
+
"step": 69000
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 18.4,
|
| 846 |
+
"learning_rate": 4.8019151345590197e-05,
|
| 847 |
+
"loss": 1.0991,
|
| 848 |
+
"step": 69500
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 18.53,
|
| 852 |
+
"learning_rate": 4.8002531308286706e-05,
|
| 853 |
+
"loss": 1.0941,
|
| 854 |
+
"step": 70000
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 18.66,
|
| 858 |
+
"learning_rate": 4.7985877964295234e-05,
|
| 859 |
+
"loss": 1.0892,
|
| 860 |
+
"step": 70500
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 18.79,
|
| 864 |
+
"learning_rate": 4.796922462030376e-05,
|
| 865 |
+
"loss": 1.0847,
|
| 866 |
+
"step": 71000
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 18.93,
|
| 870 |
+
"learning_rate": 4.7952571276312284e-05,
|
| 871 |
+
"loss": 1.083,
|
| 872 |
+
"step": 71500
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 19.06,
|
| 876 |
+
"learning_rate": 4.793591793232081e-05,
|
| 877 |
+
"loss": 1.0802,
|
| 878 |
+
"step": 72000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 19.19,
|
| 882 |
+
"learning_rate": 4.791926458832934e-05,
|
| 883 |
+
"loss": 1.0761,
|
| 884 |
+
"step": 72500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 19.32,
|
| 888 |
+
"learning_rate": 4.790261124433786e-05,
|
| 889 |
+
"loss": 1.0726,
|
| 890 |
+
"step": 73000
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 19.45,
|
| 894 |
+
"learning_rate": 4.788595790034639e-05,
|
| 895 |
+
"loss": 1.0683,
|
| 896 |
+
"step": 73500
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 19.59,
|
| 900 |
+
"learning_rate": 4.78693378630429e-05,
|
| 901 |
+
"loss": 1.0667,
|
| 902 |
+
"step": 74000
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 19.72,
|
| 906 |
+
"learning_rate": 4.785268451905143e-05,
|
| 907 |
+
"loss": 1.0655,
|
| 908 |
+
"step": 74500
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 19.85,
|
| 912 |
+
"learning_rate": 4.783603117505995e-05,
|
| 913 |
+
"loss": 1.0617,
|
| 914 |
+
"step": 75000
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 19.98,
|
| 918 |
+
"learning_rate": 4.781937783106848e-05,
|
| 919 |
+
"loss": 1.0599,
|
| 920 |
+
"step": 75500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 20.12,
|
| 924 |
+
"learning_rate": 4.780272448707701e-05,
|
| 925 |
+
"loss": 1.0554,
|
| 926 |
+
"step": 76000
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 20.25,
|
| 930 |
+
"learning_rate": 4.778607114308553e-05,
|
| 931 |
+
"loss": 1.0541,
|
| 932 |
+
"step": 76500
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 20.38,
|
| 936 |
+
"learning_rate": 4.7769417799094064e-05,
|
| 937 |
+
"loss": 1.0482,
|
| 938 |
+
"step": 77000
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 20.51,
|
| 942 |
+
"learning_rate": 4.7752764455102586e-05,
|
| 943 |
+
"loss": 1.0486,
|
| 944 |
+
"step": 77500
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 20.65,
|
| 948 |
+
"learning_rate": 4.7736144417799095e-05,
|
| 949 |
+
"loss": 1.0453,
|
| 950 |
+
"step": 78000
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 20.78,
|
| 954 |
+
"learning_rate": 4.7719491073807623e-05,
|
| 955 |
+
"loss": 1.0427,
|
| 956 |
+
"step": 78500
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 20.91,
|
| 960 |
+
"learning_rate": 4.770287103650413e-05,
|
| 961 |
+
"loss": 1.0399,
|
| 962 |
+
"step": 79000
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 21.04,
|
| 966 |
+
"learning_rate": 4.768621769251266e-05,
|
| 967 |
+
"loss": 1.0363,
|
| 968 |
+
"step": 79500
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 21.18,
|
| 972 |
+
"learning_rate": 4.766956434852118e-05,
|
| 973 |
+
"loss": 1.0321,
|
| 974 |
+
"step": 80000
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 21.31,
|
| 978 |
+
"learning_rate": 4.765291100452971e-05,
|
| 979 |
+
"loss": 1.0311,
|
| 980 |
+
"step": 80500
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 21.44,
|
| 984 |
+
"learning_rate": 4.763629096722622e-05,
|
| 985 |
+
"loss": 1.0287,
|
| 986 |
+
"step": 81000
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 21.57,
|
| 990 |
+
"learning_rate": 4.761963762323475e-05,
|
| 991 |
+
"loss": 1.0263,
|
| 992 |
+
"step": 81500
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 21.7,
|
| 996 |
+
"learning_rate": 4.760298427924327e-05,
|
| 997 |
+
"loss": 1.0241,
|
| 998 |
+
"step": 82000
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 21.84,
|
| 1002 |
+
"learning_rate": 4.75863309352518e-05,
|
| 1003 |
+
"loss": 1.0231,
|
| 1004 |
+
"step": 82500
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 21.97,
|
| 1008 |
+
"learning_rate": 4.756967759126033e-05,
|
| 1009 |
+
"loss": 1.0223,
|
| 1010 |
+
"step": 83000
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 22.1,
|
| 1014 |
+
"learning_rate": 4.7553057553956836e-05,
|
| 1015 |
+
"loss": 1.019,
|
| 1016 |
+
"step": 83500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 22.23,
|
| 1020 |
+
"learning_rate": 4.7536404209965365e-05,
|
| 1021 |
+
"loss": 1.0148,
|
| 1022 |
+
"step": 84000
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 22.37,
|
| 1026 |
+
"learning_rate": 4.7519750865973887e-05,
|
| 1027 |
+
"loss": 1.012,
|
| 1028 |
+
"step": 84500
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 22.5,
|
| 1032 |
+
"learning_rate": 4.7503097521982415e-05,
|
| 1033 |
+
"loss": 1.0117,
|
| 1034 |
+
"step": 85000
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 22.63,
|
| 1038 |
+
"learning_rate": 4.7486477484678924e-05,
|
| 1039 |
+
"loss": 1.0089,
|
| 1040 |
+
"step": 85500
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 22.76,
|
| 1044 |
+
"learning_rate": 4.746982414068745e-05,
|
| 1045 |
+
"loss": 1.0068,
|
| 1046 |
+
"step": 86000
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 22.9,
|
| 1050 |
+
"learning_rate": 4.745317079669598e-05,
|
| 1051 |
+
"loss": 1.0042,
|
| 1052 |
+
"step": 86500
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 23.03,
|
| 1056 |
+
"learning_rate": 4.74365174527045e-05,
|
| 1057 |
+
"loss": 1.002,
|
| 1058 |
+
"step": 87000
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 23.16,
|
| 1062 |
+
"learning_rate": 4.741989741540102e-05,
|
| 1063 |
+
"loss": 1.0018,
|
| 1064 |
+
"step": 87500
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 23.29,
|
| 1068 |
+
"learning_rate": 4.740324407140954e-05,
|
| 1069 |
+
"loss": 0.996,
|
| 1070 |
+
"step": 88000
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 23.43,
|
| 1074 |
+
"learning_rate": 4.738659072741807e-05,
|
| 1075 |
+
"loss": 0.9964,
|
| 1076 |
+
"step": 88500
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 23.56,
|
| 1080 |
+
"learning_rate": 4.736993738342659e-05,
|
| 1081 |
+
"loss": 0.9921,
|
| 1082 |
+
"step": 89000
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 23.69,
|
| 1086 |
+
"learning_rate": 4.735328403943512e-05,
|
| 1087 |
+
"loss": 0.9919,
|
| 1088 |
+
"step": 89500
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 23.82,
|
| 1092 |
+
"learning_rate": 4.733666400213163e-05,
|
| 1093 |
+
"loss": 0.9894,
|
| 1094 |
+
"step": 90000
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 23.95,
|
| 1098 |
+
"learning_rate": 4.7320010658140156e-05,
|
| 1099 |
+
"loss": 0.9867,
|
| 1100 |
+
"step": 90500
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 24.09,
|
| 1104 |
+
"learning_rate": 4.7303357314148685e-05,
|
| 1105 |
+
"loss": 0.9856,
|
| 1106 |
+
"step": 91000
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 24.22,
|
| 1110 |
+
"learning_rate": 4.7286737276845194e-05,
|
| 1111 |
+
"loss": 0.9821,
|
| 1112 |
+
"step": 91500
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 24.35,
|
| 1116 |
+
"learning_rate": 4.727008393285372e-05,
|
| 1117 |
+
"loss": 0.983,
|
| 1118 |
+
"step": 92000
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 24.48,
|
| 1122 |
+
"learning_rate": 4.7253430588862244e-05,
|
| 1123 |
+
"loss": 0.9782,
|
| 1124 |
+
"step": 92500
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 24.62,
|
| 1128 |
+
"learning_rate": 4.723677724487077e-05,
|
| 1129 |
+
"loss": 0.9789,
|
| 1130 |
+
"step": 93000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 24.75,
|
| 1134 |
+
"learning_rate": 4.72201239008793e-05,
|
| 1135 |
+
"loss": 0.9767,
|
| 1136 |
+
"step": 93500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 24.88,
|
| 1140 |
+
"learning_rate": 4.720347055688782e-05,
|
| 1141 |
+
"loss": 0.9741,
|
| 1142 |
+
"step": 94000
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 25.01,
|
| 1146 |
+
"learning_rate": 4.718681721289635e-05,
|
| 1147 |
+
"loss": 0.9723,
|
| 1148 |
+
"step": 94500
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 25.15,
|
| 1152 |
+
"learning_rate": 4.717016386890488e-05,
|
| 1153 |
+
"loss": 0.9688,
|
| 1154 |
+
"step": 95000
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 25.28,
|
| 1158 |
+
"learning_rate": 4.71535105249134e-05,
|
| 1159 |
+
"loss": 0.9683,
|
| 1160 |
+
"step": 95500
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 25.41,
|
| 1164 |
+
"learning_rate": 4.713689048760992e-05,
|
| 1165 |
+
"loss": 0.9683,
|
| 1166 |
+
"step": 96000
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 25.54,
|
| 1170 |
+
"learning_rate": 4.712023714361844e-05,
|
| 1171 |
+
"loss": 0.9663,
|
| 1172 |
+
"step": 96500
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 25.67,
|
| 1176 |
+
"learning_rate": 4.710358379962697e-05,
|
| 1177 |
+
"loss": 0.9623,
|
| 1178 |
+
"step": 97000
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 25.81,
|
| 1182 |
+
"learning_rate": 4.708693045563549e-05,
|
| 1183 |
+
"loss": 0.9614,
|
| 1184 |
+
"step": 97500
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 25.94,
|
| 1188 |
+
"learning_rate": 4.7070310418332005e-05,
|
| 1189 |
+
"loss": 0.9612,
|
| 1190 |
+
"step": 98000
|
| 1191 |
+
}
|
| 1192 |
+
],
|
| 1193 |
+
"max_steps": 1511200,
|
| 1194 |
+
"num_train_epochs": 400,
|
| 1195 |
+
"total_flos": 2.647131420139664e+19,
|
| 1196 |
+
"trial_name": null,
|
| 1197 |
+
"trial_params": null
|
| 1198 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ecc32ef76cdc5fb60b7231c14f207b663f286a94c08e156a88fce6a2011aea7c
|
| 3 |
+
size 3503
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|