Upload folder using huggingface_hub
Browse files- model.safetensors +1 -1
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +3 -21
- tokenizer_config.json +0 -4
- trainer_state.json +1713 -0
- training_args.bin +1 -1
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 891644712
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42c49b62d0a65de2d959d451c894f48fe178993fa432af34a0170ca34bbbde4e
|
| 3 |
size 891644712
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b71ca14cd4df9654a5995f237536e0f2335e5c5f80612e82306ad205f3e0b0d5
|
| 3 |
+
size 1783448843
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3b099612d53f6e080ee943f6f5aba2d8b05489aa0998058f89b01063c8a22084
|
| 3 |
+
size 14645
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f46db158a9baf4f3465fb8e0053ba6c6379b170fd5e4625defe29ec3007b053b
|
| 3 |
+
size 1465
|
special_tokens_map.json
CHANGED
|
@@ -101,25 +101,7 @@
|
|
| 101 |
"<extra_id_98>",
|
| 102 |
"<extra_id_99>"
|
| 103 |
],
|
| 104 |
-
"eos_token":
|
| 105 |
-
|
| 106 |
-
|
| 107 |
-
"normalized": false,
|
| 108 |
-
"rstrip": false,
|
| 109 |
-
"single_word": false
|
| 110 |
-
},
|
| 111 |
-
"pad_token": {
|
| 112 |
-
"content": "<pad>",
|
| 113 |
-
"lstrip": false,
|
| 114 |
-
"normalized": false,
|
| 115 |
-
"rstrip": false,
|
| 116 |
-
"single_word": false
|
| 117 |
-
},
|
| 118 |
-
"unk_token": {
|
| 119 |
-
"content": "<unk>",
|
| 120 |
-
"lstrip": false,
|
| 121 |
-
"normalized": false,
|
| 122 |
-
"rstrip": false,
|
| 123 |
-
"single_word": false
|
| 124 |
-
}
|
| 125 |
}
|
|
|
|
| 101 |
"<extra_id_98>",
|
| 102 |
"<extra_id_99>"
|
| 103 |
],
|
| 104 |
+
"eos_token": "</s>",
|
| 105 |
+
"pad_token": "<pad>",
|
| 106 |
+
"unk_token": "<unk>"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 107 |
}
|
tokenizer_config.json
CHANGED
|
@@ -932,12 +932,8 @@
|
|
| 932 |
"eos_token": "</s>",
|
| 933 |
"extra_ids": 100,
|
| 934 |
"extra_special_tokens": {},
|
| 935 |
-
"max_length": 64,
|
| 936 |
"model_max_length": 1000000000000000019884624838656,
|
| 937 |
"pad_token": "<pad>",
|
| 938 |
-
"stride": 0,
|
| 939 |
"tokenizer_class": "T5Tokenizer",
|
| 940 |
-
"truncation_side": "right",
|
| 941 |
-
"truncation_strategy": "longest_first",
|
| 942 |
"unk_token": "<unk>"
|
| 943 |
}
|
|
|
|
| 932 |
"eos_token": "</s>",
|
| 933 |
"extra_ids": 100,
|
| 934 |
"extra_special_tokens": {},
|
|
|
|
| 935 |
"model_max_length": 1000000000000000019884624838656,
|
| 936 |
"pad_token": "<pad>",
|
|
|
|
| 937 |
"tokenizer_class": "T5Tokenizer",
|
|
|
|
|
|
|
| 938 |
"unk_token": "<unk>"
|
| 939 |
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1713 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 11851,
|
| 3 |
+
"best_metric": 0.35250081833587826,
|
| 4 |
+
"best_model_checkpoint": "runs/t5_base/checkpoint-11851",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 11851,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.004219053244451945,
|
| 14 |
+
"grad_norm": 2.2585606575012207,
|
| 15 |
+
"learning_rate": 2.9997519196692262e-05,
|
| 16 |
+
"loss": 2.5128,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.00843810648890389,
|
| 21 |
+
"grad_norm": 2.0266594886779785,
|
| 22 |
+
"learning_rate": 2.999498776474559e-05,
|
| 23 |
+
"loss": 2.1371,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.012657159733355836,
|
| 28 |
+
"grad_norm": 2.052781105041504,
|
| 29 |
+
"learning_rate": 2.999245633279892e-05,
|
| 30 |
+
"loss": 2.1719,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.01687621297780778,
|
| 35 |
+
"grad_norm": 2.0995216369628906,
|
| 36 |
+
"learning_rate": 2.9989924900852246e-05,
|
| 37 |
+
"loss": 2.1226,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.021095266222259726,
|
| 42 |
+
"grad_norm": 2.0771021842956543,
|
| 43 |
+
"learning_rate": 2.998739346890558e-05,
|
| 44 |
+
"loss": 2.0941,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.02531431946671167,
|
| 49 |
+
"grad_norm": 1.999500036239624,
|
| 50 |
+
"learning_rate": 2.9984862036958908e-05,
|
| 51 |
+
"loss": 2.0855,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.029533372711163616,
|
| 56 |
+
"grad_norm": 2.038130283355713,
|
| 57 |
+
"learning_rate": 2.9982330605012237e-05,
|
| 58 |
+
"loss": 2.0778,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.03375242595561556,
|
| 63 |
+
"grad_norm": 1.7876685857772827,
|
| 64 |
+
"learning_rate": 2.9979799173065563e-05,
|
| 65 |
+
"loss": 2.0958,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.03797147920006751,
|
| 70 |
+
"grad_norm": 2.085517644882202,
|
| 71 |
+
"learning_rate": 2.9977267741118892e-05,
|
| 72 |
+
"loss": 2.0651,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.04219053244451945,
|
| 77 |
+
"grad_norm": 2.0681469440460205,
|
| 78 |
+
"learning_rate": 2.9974736309172224e-05,
|
| 79 |
+
"loss": 2.0345,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.0464095856889714,
|
| 84 |
+
"grad_norm": 2.10701060295105,
|
| 85 |
+
"learning_rate": 2.9972204877225554e-05,
|
| 86 |
+
"loss": 2.0222,
|
| 87 |
+
"step": 550
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.05062863893342334,
|
| 91 |
+
"grad_norm": 1.9571948051452637,
|
| 92 |
+
"learning_rate": 2.996967344527888e-05,
|
| 93 |
+
"loss": 2.02,
|
| 94 |
+
"step": 600
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.05484769217787529,
|
| 98 |
+
"grad_norm": 1.9567148685455322,
|
| 99 |
+
"learning_rate": 2.996714201333221e-05,
|
| 100 |
+
"loss": 2.0468,
|
| 101 |
+
"step": 650
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.05906674542232723,
|
| 105 |
+
"grad_norm": 1.9537338018417358,
|
| 106 |
+
"learning_rate": 2.9964610581385538e-05,
|
| 107 |
+
"loss": 2.0031,
|
| 108 |
+
"step": 700
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.06328579866677918,
|
| 112 |
+
"grad_norm": 1.8682233095169067,
|
| 113 |
+
"learning_rate": 2.9962079149438867e-05,
|
| 114 |
+
"loss": 1.9952,
|
| 115 |
+
"step": 750
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.06750485191123112,
|
| 119 |
+
"grad_norm": 2.0501787662506104,
|
| 120 |
+
"learning_rate": 2.9959547717492196e-05,
|
| 121 |
+
"loss": 1.9819,
|
| 122 |
+
"step": 800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.07172390515568307,
|
| 126 |
+
"grad_norm": 2.2852847576141357,
|
| 127 |
+
"learning_rate": 2.9957016285545525e-05,
|
| 128 |
+
"loss": 2.0428,
|
| 129 |
+
"step": 850
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.07594295840013501,
|
| 133 |
+
"grad_norm": 2.0051698684692383,
|
| 134 |
+
"learning_rate": 2.9954484853598854e-05,
|
| 135 |
+
"loss": 1.9809,
|
| 136 |
+
"step": 900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.08016201164458696,
|
| 140 |
+
"grad_norm": 1.9543101787567139,
|
| 141 |
+
"learning_rate": 2.995195342165218e-05,
|
| 142 |
+
"loss": 1.9958,
|
| 143 |
+
"step": 950
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.0843810648890389,
|
| 147 |
+
"grad_norm": 2.026174545288086,
|
| 148 |
+
"learning_rate": 2.9949421989705512e-05,
|
| 149 |
+
"loss": 1.9996,
|
| 150 |
+
"step": 1000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.08860011813349085,
|
| 154 |
+
"grad_norm": 1.820539116859436,
|
| 155 |
+
"learning_rate": 2.994689055775884e-05,
|
| 156 |
+
"loss": 1.9546,
|
| 157 |
+
"step": 1050
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.0928191713779428,
|
| 161 |
+
"grad_norm": 1.8043129444122314,
|
| 162 |
+
"learning_rate": 2.9944359125812167e-05,
|
| 163 |
+
"loss": 2.0039,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.09703822462239474,
|
| 168 |
+
"grad_norm": 2.6750881671905518,
|
| 169 |
+
"learning_rate": 2.9941827693865496e-05,
|
| 170 |
+
"loss": 1.9502,
|
| 171 |
+
"step": 1150
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.10125727786684668,
|
| 175 |
+
"grad_norm": 2.0512490272521973,
|
| 176 |
+
"learning_rate": 2.9939296261918825e-05,
|
| 177 |
+
"loss": 1.9407,
|
| 178 |
+
"step": 1200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.10547633111129863,
|
| 182 |
+
"grad_norm": 2.10312557220459,
|
| 183 |
+
"learning_rate": 2.9936764829972158e-05,
|
| 184 |
+
"loss": 1.9888,
|
| 185 |
+
"step": 1250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.10969538435575057,
|
| 189 |
+
"grad_norm": 1.769840955734253,
|
| 190 |
+
"learning_rate": 2.9934233398025484e-05,
|
| 191 |
+
"loss": 2.036,
|
| 192 |
+
"step": 1300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.11391443760020252,
|
| 196 |
+
"grad_norm": 1.9376733303070068,
|
| 197 |
+
"learning_rate": 2.9931701966078813e-05,
|
| 198 |
+
"loss": 1.9417,
|
| 199 |
+
"step": 1350
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.11813349084465447,
|
| 203 |
+
"grad_norm": 1.8222594261169434,
|
| 204 |
+
"learning_rate": 2.9929170534132142e-05,
|
| 205 |
+
"loss": 1.9711,
|
| 206 |
+
"step": 1400
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.12235254408910641,
|
| 210 |
+
"grad_norm": 2.9674904346466064,
|
| 211 |
+
"learning_rate": 2.9926639102185468e-05,
|
| 212 |
+
"loss": 1.8932,
|
| 213 |
+
"step": 1450
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.12657159733355836,
|
| 217 |
+
"grad_norm": 2.0872902870178223,
|
| 218 |
+
"learning_rate": 2.99241076702388e-05,
|
| 219 |
+
"loss": 1.9733,
|
| 220 |
+
"step": 1500
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.1307906505780103,
|
| 224 |
+
"grad_norm": 2.122377395629883,
|
| 225 |
+
"learning_rate": 2.992157623829213e-05,
|
| 226 |
+
"loss": 1.9265,
|
| 227 |
+
"step": 1550
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.13500970382246225,
|
| 231 |
+
"grad_norm": 2.2321829795837402,
|
| 232 |
+
"learning_rate": 2.991904480634546e-05,
|
| 233 |
+
"loss": 1.9439,
|
| 234 |
+
"step": 1600
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.13922875706691418,
|
| 238 |
+
"grad_norm": 10.14115047454834,
|
| 239 |
+
"learning_rate": 2.9916513374398784e-05,
|
| 240 |
+
"loss": 1.9726,
|
| 241 |
+
"step": 1650
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.14344781031136614,
|
| 245 |
+
"grad_norm": 1.6735061407089233,
|
| 246 |
+
"learning_rate": 2.9913981942452113e-05,
|
| 247 |
+
"loss": 1.9109,
|
| 248 |
+
"step": 1700
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.14766686355581807,
|
| 252 |
+
"grad_norm": 2.1993138790130615,
|
| 253 |
+
"learning_rate": 2.9911450510505446e-05,
|
| 254 |
+
"loss": 1.9683,
|
| 255 |
+
"step": 1750
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.15188591680027003,
|
| 259 |
+
"grad_norm": 1.8562805652618408,
|
| 260 |
+
"learning_rate": 2.990891907855877e-05,
|
| 261 |
+
"loss": 1.9464,
|
| 262 |
+
"step": 1800
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.15610497004472196,
|
| 266 |
+
"grad_norm": 1.7734477519989014,
|
| 267 |
+
"learning_rate": 2.99063876466121e-05,
|
| 268 |
+
"loss": 1.8865,
|
| 269 |
+
"step": 1850
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.16032402328917392,
|
| 273 |
+
"grad_norm": 5.716747760772705,
|
| 274 |
+
"learning_rate": 2.990385621466543e-05,
|
| 275 |
+
"loss": 1.8729,
|
| 276 |
+
"step": 1900
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.16454307653362585,
|
| 280 |
+
"grad_norm": 1.9271905422210693,
|
| 281 |
+
"learning_rate": 2.990132478271876e-05,
|
| 282 |
+
"loss": 1.993,
|
| 283 |
+
"step": 1950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.1687621297780778,
|
| 287 |
+
"grad_norm": 1.818063735961914,
|
| 288 |
+
"learning_rate": 2.9898793350772088e-05,
|
| 289 |
+
"loss": 1.991,
|
| 290 |
+
"step": 2000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.17298118302252974,
|
| 294 |
+
"grad_norm": 1.8673155307769775,
|
| 295 |
+
"learning_rate": 2.9896261918825417e-05,
|
| 296 |
+
"loss": 1.8647,
|
| 297 |
+
"step": 2050
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.1772002362669817,
|
| 301 |
+
"grad_norm": 2.1390140056610107,
|
| 302 |
+
"learning_rate": 2.9893730486878746e-05,
|
| 303 |
+
"loss": 1.9429,
|
| 304 |
+
"step": 2100
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.18141928951143363,
|
| 308 |
+
"grad_norm": 2.3927204608917236,
|
| 309 |
+
"learning_rate": 2.9891199054932072e-05,
|
| 310 |
+
"loss": 1.9395,
|
| 311 |
+
"step": 2150
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.1856383427558856,
|
| 315 |
+
"grad_norm": 1.9789947271347046,
|
| 316 |
+
"learning_rate": 2.98886676229854e-05,
|
| 317 |
+
"loss": 1.9396,
|
| 318 |
+
"step": 2200
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.18985739600033752,
|
| 322 |
+
"grad_norm": 1.9586814641952515,
|
| 323 |
+
"learning_rate": 2.9886136191038734e-05,
|
| 324 |
+
"loss": 1.9063,
|
| 325 |
+
"step": 2250
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.19407644924478948,
|
| 329 |
+
"grad_norm": 1.9174619913101196,
|
| 330 |
+
"learning_rate": 2.9883604759092063e-05,
|
| 331 |
+
"loss": 1.916,
|
| 332 |
+
"step": 2300
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.1982955024892414,
|
| 336 |
+
"grad_norm": 1.5862284898757935,
|
| 337 |
+
"learning_rate": 2.988107332714539e-05,
|
| 338 |
+
"loss": 1.9429,
|
| 339 |
+
"step": 2350
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.20251455573369337,
|
| 343 |
+
"grad_norm": 1.7526780366897583,
|
| 344 |
+
"learning_rate": 2.9878541895198718e-05,
|
| 345 |
+
"loss": 1.9608,
|
| 346 |
+
"step": 2400
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.2067336089781453,
|
| 350 |
+
"grad_norm": 1.6990829706192017,
|
| 351 |
+
"learning_rate": 2.9876010463252047e-05,
|
| 352 |
+
"loss": 1.9239,
|
| 353 |
+
"step": 2450
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.21095266222259726,
|
| 357 |
+
"grad_norm": 1.6032485961914062,
|
| 358 |
+
"learning_rate": 2.9873479031305376e-05,
|
| 359 |
+
"loss": 1.9369,
|
| 360 |
+
"step": 2500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.2151717154670492,
|
| 364 |
+
"grad_norm": 1.8640516996383667,
|
| 365 |
+
"learning_rate": 2.9870947599358705e-05,
|
| 366 |
+
"loss": 1.9112,
|
| 367 |
+
"step": 2550
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.21939076871150115,
|
| 371 |
+
"grad_norm": 1.777523398399353,
|
| 372 |
+
"learning_rate": 2.9868416167412034e-05,
|
| 373 |
+
"loss": 1.9128,
|
| 374 |
+
"step": 2600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.22360982195595308,
|
| 378 |
+
"grad_norm": 1.9761775732040405,
|
| 379 |
+
"learning_rate": 2.9865884735465363e-05,
|
| 380 |
+
"loss": 1.8996,
|
| 381 |
+
"step": 2650
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.22782887520040504,
|
| 385 |
+
"grad_norm": 2.1132357120513916,
|
| 386 |
+
"learning_rate": 2.986335330351869e-05,
|
| 387 |
+
"loss": 1.923,
|
| 388 |
+
"step": 2700
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.23204792844485697,
|
| 392 |
+
"grad_norm": 1.8450723886489868,
|
| 393 |
+
"learning_rate": 2.986082187157202e-05,
|
| 394 |
+
"loss": 1.9159,
|
| 395 |
+
"step": 2750
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.23626698168930893,
|
| 399 |
+
"grad_norm": 1.8609806299209595,
|
| 400 |
+
"learning_rate": 2.985829043962535e-05,
|
| 401 |
+
"loss": 1.901,
|
| 402 |
+
"step": 2800
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.24048603493376086,
|
| 406 |
+
"grad_norm": 2.161712169647217,
|
| 407 |
+
"learning_rate": 2.9855759007678676e-05,
|
| 408 |
+
"loss": 1.8938,
|
| 409 |
+
"step": 2850
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.24470508817821282,
|
| 413 |
+
"grad_norm": 1.7963075637817383,
|
| 414 |
+
"learning_rate": 2.9853227575732006e-05,
|
| 415 |
+
"loss": 1.8968,
|
| 416 |
+
"step": 2900
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.24892414142266475,
|
| 420 |
+
"grad_norm": 1.7915133237838745,
|
| 421 |
+
"learning_rate": 2.9850696143785335e-05,
|
| 422 |
+
"loss": 1.895,
|
| 423 |
+
"step": 2950
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.2531431946671167,
|
| 427 |
+
"grad_norm": 1.8958337306976318,
|
| 428 |
+
"learning_rate": 2.9848164711838667e-05,
|
| 429 |
+
"loss": 1.9464,
|
| 430 |
+
"step": 3000
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.25736224791156864,
|
| 434 |
+
"grad_norm": 1.6896766424179077,
|
| 435 |
+
"learning_rate": 2.9845633279891993e-05,
|
| 436 |
+
"loss": 1.931,
|
| 437 |
+
"step": 3050
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.2615813011560206,
|
| 441 |
+
"grad_norm": 1.745676875114441,
|
| 442 |
+
"learning_rate": 2.9843101847945322e-05,
|
| 443 |
+
"loss": 1.9099,
|
| 444 |
+
"step": 3100
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.26580035440047256,
|
| 448 |
+
"grad_norm": 2.3304221630096436,
|
| 449 |
+
"learning_rate": 2.984057041599865e-05,
|
| 450 |
+
"loss": 1.892,
|
| 451 |
+
"step": 3150
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.2700194076449245,
|
| 455 |
+
"grad_norm": 1.983554482460022,
|
| 456 |
+
"learning_rate": 2.9838038984051977e-05,
|
| 457 |
+
"loss": 1.8909,
|
| 458 |
+
"step": 3200
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.2742384608893764,
|
| 462 |
+
"grad_norm": 1.8137835264205933,
|
| 463 |
+
"learning_rate": 2.9835507552105306e-05,
|
| 464 |
+
"loss": 1.8802,
|
| 465 |
+
"step": 3250
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.27845751413382835,
|
| 469 |
+
"grad_norm": 1.5772160291671753,
|
| 470 |
+
"learning_rate": 2.983297612015864e-05,
|
| 471 |
+
"loss": 1.9312,
|
| 472 |
+
"step": 3300
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.28267656737828034,
|
| 476 |
+
"grad_norm": 1.6805251836776733,
|
| 477 |
+
"learning_rate": 2.9830444688211968e-05,
|
| 478 |
+
"loss": 1.85,
|
| 479 |
+
"step": 3350
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.2868956206227323,
|
| 483 |
+
"grad_norm": 2.302870035171509,
|
| 484 |
+
"learning_rate": 2.9827913256265293e-05,
|
| 485 |
+
"loss": 1.8832,
|
| 486 |
+
"step": 3400
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.2911146738671842,
|
| 490 |
+
"grad_norm": 1.6704341173171997,
|
| 491 |
+
"learning_rate": 2.9825381824318623e-05,
|
| 492 |
+
"loss": 1.916,
|
| 493 |
+
"step": 3450
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.29533372711163614,
|
| 497 |
+
"grad_norm": 1.758957862854004,
|
| 498 |
+
"learning_rate": 2.982285039237195e-05,
|
| 499 |
+
"loss": 1.8631,
|
| 500 |
+
"step": 3500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.29955278035608807,
|
| 504 |
+
"grad_norm": 1.78660249710083,
|
| 505 |
+
"learning_rate": 2.982031896042528e-05,
|
| 506 |
+
"loss": 1.9041,
|
| 507 |
+
"step": 3550
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.30377183360054005,
|
| 511 |
+
"grad_norm": 1.7145719528198242,
|
| 512 |
+
"learning_rate": 2.981778752847861e-05,
|
| 513 |
+
"loss": 1.8948,
|
| 514 |
+
"step": 3600
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.307990886844992,
|
| 518 |
+
"grad_norm": 1.6444560289382935,
|
| 519 |
+
"learning_rate": 2.981525609653194e-05,
|
| 520 |
+
"loss": 1.8902,
|
| 521 |
+
"step": 3650
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.3122099400894439,
|
| 525 |
+
"grad_norm": 1.5642787218093872,
|
| 526 |
+
"learning_rate": 2.9812724664585268e-05,
|
| 527 |
+
"loss": 1.9288,
|
| 528 |
+
"step": 3700
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.31642899333389585,
|
| 532 |
+
"grad_norm": 1.8566147089004517,
|
| 533 |
+
"learning_rate": 2.9810193232638594e-05,
|
| 534 |
+
"loss": 1.8711,
|
| 535 |
+
"step": 3750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.32064804657834783,
|
| 539 |
+
"grad_norm": 1.7470507621765137,
|
| 540 |
+
"learning_rate": 2.9807661800691926e-05,
|
| 541 |
+
"loss": 1.8988,
|
| 542 |
+
"step": 3800
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.32486709982279977,
|
| 546 |
+
"grad_norm": 2.019124984741211,
|
| 547 |
+
"learning_rate": 2.9805130368745256e-05,
|
| 548 |
+
"loss": 1.8682,
|
| 549 |
+
"step": 3850
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.3290861530672517,
|
| 553 |
+
"grad_norm": 1.7901630401611328,
|
| 554 |
+
"learning_rate": 2.980259893679858e-05,
|
| 555 |
+
"loss": 1.9112,
|
| 556 |
+
"step": 3900
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.33330520631170363,
|
| 560 |
+
"grad_norm": 1.8826500177383423,
|
| 561 |
+
"learning_rate": 2.980006750485191e-05,
|
| 562 |
+
"loss": 1.885,
|
| 563 |
+
"step": 3950
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.3375242595561556,
|
| 567 |
+
"grad_norm": 1.5955331325531006,
|
| 568 |
+
"learning_rate": 2.979753607290524e-05,
|
| 569 |
+
"loss": 1.9354,
|
| 570 |
+
"step": 4000
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.34174331280060755,
|
| 574 |
+
"grad_norm": 1.7695971727371216,
|
| 575 |
+
"learning_rate": 2.9795004640958572e-05,
|
| 576 |
+
"loss": 1.8961,
|
| 577 |
+
"step": 4050
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.3459623660450595,
|
| 581 |
+
"grad_norm": 1.7151463031768799,
|
| 582 |
+
"learning_rate": 2.9792473209011898e-05,
|
| 583 |
+
"loss": 1.9173,
|
| 584 |
+
"step": 4100
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.3501814192895114,
|
| 588 |
+
"grad_norm": 1.9178495407104492,
|
| 589 |
+
"learning_rate": 2.9789941777065227e-05,
|
| 590 |
+
"loss": 1.874,
|
| 591 |
+
"step": 4150
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.3544004725339634,
|
| 595 |
+
"grad_norm": 1.8744202852249146,
|
| 596 |
+
"learning_rate": 2.9787410345118556e-05,
|
| 597 |
+
"loss": 1.8789,
|
| 598 |
+
"step": 4200
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.3586195257784153,
|
| 602 |
+
"grad_norm": 2.3288185596466064,
|
| 603 |
+
"learning_rate": 2.9784878913171885e-05,
|
| 604 |
+
"loss": 1.876,
|
| 605 |
+
"step": 4250
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.36283857902286726,
|
| 609 |
+
"grad_norm": 1.6763617992401123,
|
| 610 |
+
"learning_rate": 2.9782347481225214e-05,
|
| 611 |
+
"loss": 1.9079,
|
| 612 |
+
"step": 4300
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.3670576322673192,
|
| 616 |
+
"grad_norm": 1.5563552379608154,
|
| 617 |
+
"learning_rate": 2.9779816049278543e-05,
|
| 618 |
+
"loss": 1.9238,
|
| 619 |
+
"step": 4350
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.3712766855117712,
|
| 623 |
+
"grad_norm": 1.7934948205947876,
|
| 624 |
+
"learning_rate": 2.9777284617331872e-05,
|
| 625 |
+
"loss": 1.9149,
|
| 626 |
+
"step": 4400
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.3754957387562231,
|
| 630 |
+
"grad_norm": 1.6888962984085083,
|
| 631 |
+
"learning_rate": 2.9774753185385198e-05,
|
| 632 |
+
"loss": 1.8906,
|
| 633 |
+
"step": 4450
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.37971479200067504,
|
| 637 |
+
"grad_norm": 1.7022356986999512,
|
| 638 |
+
"learning_rate": 2.9772221753438527e-05,
|
| 639 |
+
"loss": 1.8644,
|
| 640 |
+
"step": 4500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.38393384524512697,
|
| 644 |
+
"grad_norm": 1.6906685829162598,
|
| 645 |
+
"learning_rate": 2.976969032149186e-05,
|
| 646 |
+
"loss": 1.8859,
|
| 647 |
+
"step": 4550
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.38815289848957896,
|
| 651 |
+
"grad_norm": 1.7702606916427612,
|
| 652 |
+
"learning_rate": 2.976715888954519e-05,
|
| 653 |
+
"loss": 1.8658,
|
| 654 |
+
"step": 4600
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.3923719517340309,
|
| 658 |
+
"grad_norm": 1.8164970874786377,
|
| 659 |
+
"learning_rate": 2.9764627457598515e-05,
|
| 660 |
+
"loss": 1.873,
|
| 661 |
+
"step": 4650
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.3965910049784828,
|
| 665 |
+
"grad_norm": 1.8003426790237427,
|
| 666 |
+
"learning_rate": 2.9762096025651844e-05,
|
| 667 |
+
"loss": 1.9107,
|
| 668 |
+
"step": 4700
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.40081005822293475,
|
| 672 |
+
"grad_norm": 2.902315378189087,
|
| 673 |
+
"learning_rate": 2.9759564593705173e-05,
|
| 674 |
+
"loss": 1.8885,
|
| 675 |
+
"step": 4750
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.40502911146738674,
|
| 679 |
+
"grad_norm": 1.896249771118164,
|
| 680 |
+
"learning_rate": 2.9757033161758502e-05,
|
| 681 |
+
"loss": 1.8555,
|
| 682 |
+
"step": 4800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.40924816471183867,
|
| 686 |
+
"grad_norm": 1.8151708841323853,
|
| 687 |
+
"learning_rate": 2.975450172981183e-05,
|
| 688 |
+
"loss": 1.8569,
|
| 689 |
+
"step": 4850
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.4134672179562906,
|
| 693 |
+
"grad_norm": 1.7401193380355835,
|
| 694 |
+
"learning_rate": 2.975197029786516e-05,
|
| 695 |
+
"loss": 1.8601,
|
| 696 |
+
"step": 4900
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.41768627120074253,
|
| 700 |
+
"grad_norm": 1.6459238529205322,
|
| 701 |
+
"learning_rate": 2.974943886591849e-05,
|
| 702 |
+
"loss": 1.9279,
|
| 703 |
+
"step": 4950
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.4219053244451945,
|
| 707 |
+
"grad_norm": 1.4960553646087646,
|
| 708 |
+
"learning_rate": 2.9746907433971815e-05,
|
| 709 |
+
"loss": 1.8707,
|
| 710 |
+
"step": 5000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.42612437768964645,
|
| 714 |
+
"grad_norm": 1.7572098970413208,
|
| 715 |
+
"learning_rate": 2.9744376002025148e-05,
|
| 716 |
+
"loss": 1.8638,
|
| 717 |
+
"step": 5050
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.4303434309340984,
|
| 721 |
+
"grad_norm": 2.1447529792785645,
|
| 722 |
+
"learning_rate": 2.9741844570078477e-05,
|
| 723 |
+
"loss": 1.8834,
|
| 724 |
+
"step": 5100
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.4345624841785503,
|
| 728 |
+
"grad_norm": 1.478371500968933,
|
| 729 |
+
"learning_rate": 2.9739313138131803e-05,
|
| 730 |
+
"loss": 1.8837,
|
| 731 |
+
"step": 5150
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.4387815374230023,
|
| 735 |
+
"grad_norm": 2.460432291030884,
|
| 736 |
+
"learning_rate": 2.9736781706185132e-05,
|
| 737 |
+
"loss": 1.8691,
|
| 738 |
+
"step": 5200
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.44300059066745423,
|
| 742 |
+
"grad_norm": 1.8758426904678345,
|
| 743 |
+
"learning_rate": 2.973425027423846e-05,
|
| 744 |
+
"loss": 1.8694,
|
| 745 |
+
"step": 5250
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.44721964391190616,
|
| 749 |
+
"grad_norm": 2.034414052963257,
|
| 750 |
+
"learning_rate": 2.9731718842291793e-05,
|
| 751 |
+
"loss": 1.9235,
|
| 752 |
+
"step": 5300
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.4514386971563581,
|
| 756 |
+
"grad_norm": 2.1323251724243164,
|
| 757 |
+
"learning_rate": 2.972918741034512e-05,
|
| 758 |
+
"loss": 1.9031,
|
| 759 |
+
"step": 5350
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.4556577504008101,
|
| 763 |
+
"grad_norm": 1.9008245468139648,
|
| 764 |
+
"learning_rate": 2.9726655978398448e-05,
|
| 765 |
+
"loss": 1.8701,
|
| 766 |
+
"step": 5400
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.459876803645262,
|
| 770 |
+
"grad_norm": 1.8013249635696411,
|
| 771 |
+
"learning_rate": 2.9724124546451777e-05,
|
| 772 |
+
"loss": 1.8683,
|
| 773 |
+
"step": 5450
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.46409585688971394,
|
| 777 |
+
"grad_norm": 1.7143794298171997,
|
| 778 |
+
"learning_rate": 2.9721593114505103e-05,
|
| 779 |
+
"loss": 1.864,
|
| 780 |
+
"step": 5500
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.4683149101341659,
|
| 784 |
+
"grad_norm": 1.7622582912445068,
|
| 785 |
+
"learning_rate": 2.9719061682558436e-05,
|
| 786 |
+
"loss": 1.8849,
|
| 787 |
+
"step": 5550
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.47253396337861786,
|
| 791 |
+
"grad_norm": 1.608937382698059,
|
| 792 |
+
"learning_rate": 2.9716530250611765e-05,
|
| 793 |
+
"loss": 1.8885,
|
| 794 |
+
"step": 5600
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.4767530166230698,
|
| 798 |
+
"grad_norm": 1.937638759613037,
|
| 799 |
+
"learning_rate": 2.9713998818665094e-05,
|
| 800 |
+
"loss": 1.8245,
|
| 801 |
+
"step": 5650
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4809720698675217,
|
| 805 |
+
"grad_norm": 1.561886191368103,
|
| 806 |
+
"learning_rate": 2.971146738671842e-05,
|
| 807 |
+
"loss": 1.8753,
|
| 808 |
+
"step": 5700
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.48519112311197365,
|
| 812 |
+
"grad_norm": 1.684636116027832,
|
| 813 |
+
"learning_rate": 2.970893595477175e-05,
|
| 814 |
+
"loss": 1.921,
|
| 815 |
+
"step": 5750
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.48941017635642564,
|
| 819 |
+
"grad_norm": 1.8537026643753052,
|
| 820 |
+
"learning_rate": 2.970640452282508e-05,
|
| 821 |
+
"loss": 1.8571,
|
| 822 |
+
"step": 5800
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.4936292296008776,
|
| 826 |
+
"grad_norm": 1.6456936597824097,
|
| 827 |
+
"learning_rate": 2.9703873090878407e-05,
|
| 828 |
+
"loss": 1.8802,
|
| 829 |
+
"step": 5850
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.4978482828453295,
|
| 833 |
+
"grad_norm": 5.627963542938232,
|
| 834 |
+
"learning_rate": 2.9701341658931736e-05,
|
| 835 |
+
"loss": 1.8922,
|
| 836 |
+
"step": 5900
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.5020673360897815,
|
| 840 |
+
"grad_norm": 1.9793373346328735,
|
| 841 |
+
"learning_rate": 2.9698810226985065e-05,
|
| 842 |
+
"loss": 1.8128,
|
| 843 |
+
"step": 5950
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.5062863893342334,
|
| 847 |
+
"grad_norm": 1.7868801355361938,
|
| 848 |
+
"learning_rate": 2.9696278795038394e-05,
|
| 849 |
+
"loss": 1.9061,
|
| 850 |
+
"step": 6000
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.5105054425786854,
|
| 854 |
+
"grad_norm": 1.8355735540390015,
|
| 855 |
+
"learning_rate": 2.9693747363091723e-05,
|
| 856 |
+
"loss": 1.8505,
|
| 857 |
+
"step": 6050
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.5147244958231373,
|
| 861 |
+
"grad_norm": 1.7057996988296509,
|
| 862 |
+
"learning_rate": 2.9691215931145053e-05,
|
| 863 |
+
"loss": 1.9155,
|
| 864 |
+
"step": 6100
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.5189435490675892,
|
| 868 |
+
"grad_norm": 1.9731509685516357,
|
| 869 |
+
"learning_rate": 2.9688684499198382e-05,
|
| 870 |
+
"loss": 1.9031,
|
| 871 |
+
"step": 6150
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.5231626023120411,
|
| 875 |
+
"grad_norm": 1.561681866645813,
|
| 876 |
+
"learning_rate": 2.9686153067251707e-05,
|
| 877 |
+
"loss": 1.8709,
|
| 878 |
+
"step": 6200
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.5273816555564931,
|
| 882 |
+
"grad_norm": 1.6530567407608032,
|
| 883 |
+
"learning_rate": 2.9683621635305037e-05,
|
| 884 |
+
"loss": 1.8696,
|
| 885 |
+
"step": 6250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.5316007088009451,
|
| 889 |
+
"grad_norm": 1.7582885026931763,
|
| 890 |
+
"learning_rate": 2.968109020335837e-05,
|
| 891 |
+
"loss": 1.9102,
|
| 892 |
+
"step": 6300
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.535819762045397,
|
| 896 |
+
"grad_norm": 1.8595643043518066,
|
| 897 |
+
"learning_rate": 2.9678558771411698e-05,
|
| 898 |
+
"loss": 1.8665,
|
| 899 |
+
"step": 6350
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.540038815289849,
|
| 903 |
+
"grad_norm": 1.5705363750457764,
|
| 904 |
+
"learning_rate": 2.9676027339465024e-05,
|
| 905 |
+
"loss": 1.8451,
|
| 906 |
+
"step": 6400
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.5442578685343009,
|
| 910 |
+
"grad_norm": 1.543035626411438,
|
| 911 |
+
"learning_rate": 2.9673495907518353e-05,
|
| 912 |
+
"loss": 1.8647,
|
| 913 |
+
"step": 6450
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.5484769217787528,
|
| 917 |
+
"grad_norm": 1.5700854063034058,
|
| 918 |
+
"learning_rate": 2.9670964475571682e-05,
|
| 919 |
+
"loss": 1.8877,
|
| 920 |
+
"step": 6500
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.5526959750232048,
|
| 924 |
+
"grad_norm": 1.8307514190673828,
|
| 925 |
+
"learning_rate": 2.966843304362501e-05,
|
| 926 |
+
"loss": 1.8777,
|
| 927 |
+
"step": 6550
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.5569150282676567,
|
| 931 |
+
"grad_norm": 1.8251144886016846,
|
| 932 |
+
"learning_rate": 2.966590161167834e-05,
|
| 933 |
+
"loss": 1.8222,
|
| 934 |
+
"step": 6600
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.5611340815121086,
|
| 938 |
+
"grad_norm": 1.768386960029602,
|
| 939 |
+
"learning_rate": 2.966337017973167e-05,
|
| 940 |
+
"loss": 1.8527,
|
| 941 |
+
"step": 6650
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.5653531347565607,
|
| 945 |
+
"grad_norm": 1.9443879127502441,
|
| 946 |
+
"learning_rate": 2.9660838747785e-05,
|
| 947 |
+
"loss": 1.898,
|
| 948 |
+
"step": 6700
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.5695721880010126,
|
| 952 |
+
"grad_norm": 1.8102383613586426,
|
| 953 |
+
"learning_rate": 2.9658307315838324e-05,
|
| 954 |
+
"loss": 1.911,
|
| 955 |
+
"step": 6750
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.5737912412454645,
|
| 959 |
+
"grad_norm": 1.830838680267334,
|
| 960 |
+
"learning_rate": 2.9655775883891657e-05,
|
| 961 |
+
"loss": 1.8574,
|
| 962 |
+
"step": 6800
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5780102944899165,
|
| 966 |
+
"grad_norm": 2.000922679901123,
|
| 967 |
+
"learning_rate": 2.9653244451944986e-05,
|
| 968 |
+
"loss": 1.859,
|
| 969 |
+
"step": 6850
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5822293477343684,
|
| 973 |
+
"grad_norm": 1.8055535554885864,
|
| 974 |
+
"learning_rate": 2.9650713019998312e-05,
|
| 975 |
+
"loss": 1.8651,
|
| 976 |
+
"step": 6900
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.5864484009788203,
|
| 980 |
+
"grad_norm": 2.0820086002349854,
|
| 981 |
+
"learning_rate": 2.964818158805164e-05,
|
| 982 |
+
"loss": 1.8264,
|
| 983 |
+
"step": 6950
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.5906674542232723,
|
| 987 |
+
"grad_norm": 1.6195132732391357,
|
| 988 |
+
"learning_rate": 2.964565015610497e-05,
|
| 989 |
+
"loss": 1.8664,
|
| 990 |
+
"step": 7000
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.5948865074677242,
|
| 994 |
+
"grad_norm": 1.5151090621948242,
|
| 995 |
+
"learning_rate": 2.9643118724158303e-05,
|
| 996 |
+
"loss": 1.8351,
|
| 997 |
+
"step": 7050
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.5991055607121761,
|
| 1001 |
+
"grad_norm": 1.809647560119629,
|
| 1002 |
+
"learning_rate": 2.9640587292211628e-05,
|
| 1003 |
+
"loss": 1.8224,
|
| 1004 |
+
"step": 7100
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.6033246139566282,
|
| 1008 |
+
"grad_norm": 1.6554861068725586,
|
| 1009 |
+
"learning_rate": 2.9638055860264957e-05,
|
| 1010 |
+
"loss": 1.836,
|
| 1011 |
+
"step": 7150
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.6075436672010801,
|
| 1015 |
+
"grad_norm": 1.7558064460754395,
|
| 1016 |
+
"learning_rate": 2.9635524428318287e-05,
|
| 1017 |
+
"loss": 1.8585,
|
| 1018 |
+
"step": 7200
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.611762720445532,
|
| 1022 |
+
"grad_norm": 1.5888464450836182,
|
| 1023 |
+
"learning_rate": 2.9632992996371612e-05,
|
| 1024 |
+
"loss": 1.8509,
|
| 1025 |
+
"step": 7250
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.615981773689984,
|
| 1029 |
+
"grad_norm": 1.834747076034546,
|
| 1030 |
+
"learning_rate": 2.963046156442494e-05,
|
| 1031 |
+
"loss": 1.9303,
|
| 1032 |
+
"step": 7300
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.6202008269344359,
|
| 1036 |
+
"grad_norm": 1.9495407342910767,
|
| 1037 |
+
"learning_rate": 2.9627930132478274e-05,
|
| 1038 |
+
"loss": 1.8844,
|
| 1039 |
+
"step": 7350
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.6244198801788878,
|
| 1043 |
+
"grad_norm": 1.7994617223739624,
|
| 1044 |
+
"learning_rate": 2.9625398700531603e-05,
|
| 1045 |
+
"loss": 1.8402,
|
| 1046 |
+
"step": 7400
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.6286389334233398,
|
| 1050 |
+
"grad_norm": 1.8380210399627686,
|
| 1051 |
+
"learning_rate": 2.962286726858493e-05,
|
| 1052 |
+
"loss": 1.9005,
|
| 1053 |
+
"step": 7450
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.6328579866677917,
|
| 1057 |
+
"grad_norm": 1.6115747690200806,
|
| 1058 |
+
"learning_rate": 2.9620335836638258e-05,
|
| 1059 |
+
"loss": 1.8887,
|
| 1060 |
+
"step": 7500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.6370770399122437,
|
| 1064 |
+
"grad_norm": 1.7630914449691772,
|
| 1065 |
+
"learning_rate": 2.9617804404691587e-05,
|
| 1066 |
+
"loss": 1.8077,
|
| 1067 |
+
"step": 7550
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.6412960931566957,
|
| 1071 |
+
"grad_norm": 1.9116686582565308,
|
| 1072 |
+
"learning_rate": 2.9615272972744916e-05,
|
| 1073 |
+
"loss": 1.8479,
|
| 1074 |
+
"step": 7600
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.6455151464011476,
|
| 1078 |
+
"grad_norm": 1.8529033660888672,
|
| 1079 |
+
"learning_rate": 2.9612741540798245e-05,
|
| 1080 |
+
"loss": 1.827,
|
| 1081 |
+
"step": 7650
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.6497341996455995,
|
| 1085 |
+
"grad_norm": 1.9074718952178955,
|
| 1086 |
+
"learning_rate": 2.9610210108851574e-05,
|
| 1087 |
+
"loss": 1.8143,
|
| 1088 |
+
"step": 7700
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.6539532528900515,
|
| 1092 |
+
"grad_norm": 1.555598497390747,
|
| 1093 |
+
"learning_rate": 2.9607678676904904e-05,
|
| 1094 |
+
"loss": 1.8088,
|
| 1095 |
+
"step": 7750
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.6581723061345034,
|
| 1099 |
+
"grad_norm": 1.6468048095703125,
|
| 1100 |
+
"learning_rate": 2.960514724495823e-05,
|
| 1101 |
+
"loss": 1.8892,
|
| 1102 |
+
"step": 7800
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.6623913593789553,
|
| 1106 |
+
"grad_norm": 1.8467329740524292,
|
| 1107 |
+
"learning_rate": 2.9602615813011562e-05,
|
| 1108 |
+
"loss": 1.8351,
|
| 1109 |
+
"step": 7850
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.6666104126234073,
|
| 1113 |
+
"grad_norm": 3.6700387001037598,
|
| 1114 |
+
"learning_rate": 2.960008438106489e-05,
|
| 1115 |
+
"loss": 1.8171,
|
| 1116 |
+
"step": 7900
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.6708294658678593,
|
| 1120 |
+
"grad_norm": 1.5956077575683594,
|
| 1121 |
+
"learning_rate": 2.959755294911822e-05,
|
| 1122 |
+
"loss": 1.8133,
|
| 1123 |
+
"step": 7950
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.6750485191123112,
|
| 1127 |
+
"grad_norm": 2.291576385498047,
|
| 1128 |
+
"learning_rate": 2.9595021517171546e-05,
|
| 1129 |
+
"loss": 1.9014,
|
| 1130 |
+
"step": 8000
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.6792675723567632,
|
| 1134 |
+
"grad_norm": 1.5761691331863403,
|
| 1135 |
+
"learning_rate": 2.9592490085224875e-05,
|
| 1136 |
+
"loss": 1.7983,
|
| 1137 |
+
"step": 8050
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.6834866256012151,
|
| 1141 |
+
"grad_norm": 1.9439380168914795,
|
| 1142 |
+
"learning_rate": 2.9589958653278207e-05,
|
| 1143 |
+
"loss": 1.832,
|
| 1144 |
+
"step": 8100
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.687705678845667,
|
| 1148 |
+
"grad_norm": 1.711478352546692,
|
| 1149 |
+
"learning_rate": 2.9587427221331533e-05,
|
| 1150 |
+
"loss": 1.8716,
|
| 1151 |
+
"step": 8150
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.691924732090119,
|
| 1155 |
+
"grad_norm": 1.7239052057266235,
|
| 1156 |
+
"learning_rate": 2.9584895789384862e-05,
|
| 1157 |
+
"loss": 1.8725,
|
| 1158 |
+
"step": 8200
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.6961437853345709,
|
| 1162 |
+
"grad_norm": 1.832995891571045,
|
| 1163 |
+
"learning_rate": 2.958236435743819e-05,
|
| 1164 |
+
"loss": 1.8667,
|
| 1165 |
+
"step": 8250
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.7003628385790228,
|
| 1169 |
+
"grad_norm": 1.9121317863464355,
|
| 1170 |
+
"learning_rate": 2.957983292549152e-05,
|
| 1171 |
+
"loss": 1.8333,
|
| 1172 |
+
"step": 8300
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.7045818918234749,
|
| 1176 |
+
"grad_norm": 1.7547369003295898,
|
| 1177 |
+
"learning_rate": 2.957730149354485e-05,
|
| 1178 |
+
"loss": 1.8453,
|
| 1179 |
+
"step": 8350
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.7088009450679268,
|
| 1183 |
+
"grad_norm": 2.50349497795105,
|
| 1184 |
+
"learning_rate": 2.957477006159818e-05,
|
| 1185 |
+
"loss": 1.8393,
|
| 1186 |
+
"step": 8400
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.7130199983123787,
|
| 1190 |
+
"grad_norm": 2.005995273590088,
|
| 1191 |
+
"learning_rate": 2.9572238629651508e-05,
|
| 1192 |
+
"loss": 1.8636,
|
| 1193 |
+
"step": 8450
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.7172390515568307,
|
| 1197 |
+
"grad_norm": 1.933449149131775,
|
| 1198 |
+
"learning_rate": 2.9569707197704834e-05,
|
| 1199 |
+
"loss": 1.8273,
|
| 1200 |
+
"step": 8500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.7214581048012826,
|
| 1204 |
+
"grad_norm": 1.7801495790481567,
|
| 1205 |
+
"learning_rate": 2.9567175765758163e-05,
|
| 1206 |
+
"loss": 1.8626,
|
| 1207 |
+
"step": 8550
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.7256771580457345,
|
| 1211 |
+
"grad_norm": 1.8393090963363647,
|
| 1212 |
+
"learning_rate": 2.9564644333811495e-05,
|
| 1213 |
+
"loss": 1.8684,
|
| 1214 |
+
"step": 8600
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.7298962112901864,
|
| 1218 |
+
"grad_norm": 1.7270387411117554,
|
| 1219 |
+
"learning_rate": 2.9562112901864824e-05,
|
| 1220 |
+
"loss": 1.8777,
|
| 1221 |
+
"step": 8650
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.7341152645346384,
|
| 1225 |
+
"grad_norm": 1.775639533996582,
|
| 1226 |
+
"learning_rate": 2.955958146991815e-05,
|
| 1227 |
+
"loss": 1.8497,
|
| 1228 |
+
"step": 8700
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.7383343177790904,
|
| 1232 |
+
"grad_norm": 1.5626294612884521,
|
| 1233 |
+
"learning_rate": 2.955705003797148e-05,
|
| 1234 |
+
"loss": 1.8255,
|
| 1235 |
+
"step": 8750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.7425533710235424,
|
| 1239 |
+
"grad_norm": 1.640458106994629,
|
| 1240 |
+
"learning_rate": 2.955451860602481e-05,
|
| 1241 |
+
"loss": 1.8841,
|
| 1242 |
+
"step": 8800
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.7467724242679943,
|
| 1246 |
+
"grad_norm": 1.6224232912063599,
|
| 1247 |
+
"learning_rate": 2.9551987174078138e-05,
|
| 1248 |
+
"loss": 1.8381,
|
| 1249 |
+
"step": 8850
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.7509914775124462,
|
| 1253 |
+
"grad_norm": 1.6448860168457031,
|
| 1254 |
+
"learning_rate": 2.9549455742131467e-05,
|
| 1255 |
+
"loss": 1.8452,
|
| 1256 |
+
"step": 8900
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.7552105307568981,
|
| 1260 |
+
"grad_norm": 1.8404731750488281,
|
| 1261 |
+
"learning_rate": 2.9546924310184796e-05,
|
| 1262 |
+
"loss": 1.8604,
|
| 1263 |
+
"step": 8950
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.7594295840013501,
|
| 1267 |
+
"grad_norm": 1.8648627996444702,
|
| 1268 |
+
"learning_rate": 2.9544392878238125e-05,
|
| 1269 |
+
"loss": 1.8599,
|
| 1270 |
+
"step": 9000
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.763648637245802,
|
| 1274 |
+
"grad_norm": 1.7747282981872559,
|
| 1275 |
+
"learning_rate": 2.954186144629145e-05,
|
| 1276 |
+
"loss": 1.8113,
|
| 1277 |
+
"step": 9050
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.7678676904902539,
|
| 1281 |
+
"grad_norm": 1.7520033121109009,
|
| 1282 |
+
"learning_rate": 2.9539330014344783e-05,
|
| 1283 |
+
"loss": 1.8607,
|
| 1284 |
+
"step": 9100
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.772086743734706,
|
| 1288 |
+
"grad_norm": 1.549561858177185,
|
| 1289 |
+
"learning_rate": 2.9536798582398112e-05,
|
| 1290 |
+
"loss": 1.8295,
|
| 1291 |
+
"step": 9150
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.7763057969791579,
|
| 1295 |
+
"grad_norm": 1.8901214599609375,
|
| 1296 |
+
"learning_rate": 2.9534267150451438e-05,
|
| 1297 |
+
"loss": 1.9064,
|
| 1298 |
+
"step": 9200
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.7805248502236098,
|
| 1302 |
+
"grad_norm": 1.7078797817230225,
|
| 1303 |
+
"learning_rate": 2.9531735718504767e-05,
|
| 1304 |
+
"loss": 1.811,
|
| 1305 |
+
"step": 9250
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.7847439034680618,
|
| 1309 |
+
"grad_norm": 1.8151720762252808,
|
| 1310 |
+
"learning_rate": 2.9529204286558096e-05,
|
| 1311 |
+
"loss": 1.844,
|
| 1312 |
+
"step": 9300
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.7889629567125137,
|
| 1316 |
+
"grad_norm": 1.811226487159729,
|
| 1317 |
+
"learning_rate": 2.952667285461143e-05,
|
| 1318 |
+
"loss": 1.8136,
|
| 1319 |
+
"step": 9350
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.7931820099569656,
|
| 1323 |
+
"grad_norm": 1.5015605688095093,
|
| 1324 |
+
"learning_rate": 2.9524141422664754e-05,
|
| 1325 |
+
"loss": 1.827,
|
| 1326 |
+
"step": 9400
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.7974010632014176,
|
| 1330 |
+
"grad_norm": 1.5796177387237549,
|
| 1331 |
+
"learning_rate": 2.9521609990718084e-05,
|
| 1332 |
+
"loss": 1.8379,
|
| 1333 |
+
"step": 9450
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.8016201164458695,
|
| 1337 |
+
"grad_norm": 1.8602157831192017,
|
| 1338 |
+
"learning_rate": 2.9519078558771413e-05,
|
| 1339 |
+
"loss": 1.8301,
|
| 1340 |
+
"step": 9500
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.8058391696903215,
|
| 1344 |
+
"grad_norm": 2.2987871170043945,
|
| 1345 |
+
"learning_rate": 2.951654712682474e-05,
|
| 1346 |
+
"loss": 1.8367,
|
| 1347 |
+
"step": 9550
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.8100582229347735,
|
| 1351 |
+
"grad_norm": 1.707922101020813,
|
| 1352 |
+
"learning_rate": 2.951401569487807e-05,
|
| 1353 |
+
"loss": 1.8173,
|
| 1354 |
+
"step": 9600
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.8142772761792254,
|
| 1358 |
+
"grad_norm": 1.7747459411621094,
|
| 1359 |
+
"learning_rate": 2.95114842629314e-05,
|
| 1360 |
+
"loss": 1.8379,
|
| 1361 |
+
"step": 9650
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.8184963294236773,
|
| 1365 |
+
"grad_norm": 1.6157597303390503,
|
| 1366 |
+
"learning_rate": 2.950895283098473e-05,
|
| 1367 |
+
"loss": 1.8579,
|
| 1368 |
+
"step": 9700
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.8227153826681293,
|
| 1372 |
+
"grad_norm": 1.5862922668457031,
|
| 1373 |
+
"learning_rate": 2.9506421399038055e-05,
|
| 1374 |
+
"loss": 1.8351,
|
| 1375 |
+
"step": 9750
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.8269344359125812,
|
| 1379 |
+
"grad_norm": 1.7208493947982788,
|
| 1380 |
+
"learning_rate": 2.9503889967091384e-05,
|
| 1381 |
+
"loss": 1.8394,
|
| 1382 |
+
"step": 9800
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.8311534891570331,
|
| 1386 |
+
"grad_norm": 1.5757873058319092,
|
| 1387 |
+
"learning_rate": 2.9501358535144717e-05,
|
| 1388 |
+
"loss": 1.869,
|
| 1389 |
+
"step": 9850
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.8353725424014851,
|
| 1393 |
+
"grad_norm": 1.992253065109253,
|
| 1394 |
+
"learning_rate": 2.9498827103198042e-05,
|
| 1395 |
+
"loss": 1.8449,
|
| 1396 |
+
"step": 9900
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.8395915956459371,
|
| 1400 |
+
"grad_norm": 1.8114345073699951,
|
| 1401 |
+
"learning_rate": 2.949629567125137e-05,
|
| 1402 |
+
"loss": 1.8315,
|
| 1403 |
+
"step": 9950
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.843810648890389,
|
| 1407 |
+
"grad_norm": 1.69283127784729,
|
| 1408 |
+
"learning_rate": 2.94937642393047e-05,
|
| 1409 |
+
"loss": 1.8311,
|
| 1410 |
+
"step": 10000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.848029702134841,
|
| 1414 |
+
"grad_norm": 1.7031219005584717,
|
| 1415 |
+
"learning_rate": 2.949123280735803e-05,
|
| 1416 |
+
"loss": 1.8895,
|
| 1417 |
+
"step": 10050
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.8522487553792929,
|
| 1421 |
+
"grad_norm": 1.7500584125518799,
|
| 1422 |
+
"learning_rate": 2.948870137541136e-05,
|
| 1423 |
+
"loss": 1.7881,
|
| 1424 |
+
"step": 10100
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.8564678086237448,
|
| 1428 |
+
"grad_norm": 1.9866597652435303,
|
| 1429 |
+
"learning_rate": 2.9486169943464688e-05,
|
| 1430 |
+
"loss": 1.7861,
|
| 1431 |
+
"step": 10150
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.8606868618681968,
|
| 1435 |
+
"grad_norm": 1.588376760482788,
|
| 1436 |
+
"learning_rate": 2.9483638511518017e-05,
|
| 1437 |
+
"loss": 1.7675,
|
| 1438 |
+
"step": 10200
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.8649059151126487,
|
| 1442 |
+
"grad_norm": 1.966997504234314,
|
| 1443 |
+
"learning_rate": 2.9481107079571343e-05,
|
| 1444 |
+
"loss": 1.8403,
|
| 1445 |
+
"step": 10250
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.8691249683571006,
|
| 1449 |
+
"grad_norm": 1.5755244493484497,
|
| 1450 |
+
"learning_rate": 2.9478575647624672e-05,
|
| 1451 |
+
"loss": 1.8107,
|
| 1452 |
+
"step": 10300
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.8733440216015527,
|
| 1456 |
+
"grad_norm": 1.9831382036209106,
|
| 1457 |
+
"learning_rate": 2.9476044215678004e-05,
|
| 1458 |
+
"loss": 1.8022,
|
| 1459 |
+
"step": 10350
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.8775630748460046,
|
| 1463 |
+
"grad_norm": 2.213834047317505,
|
| 1464 |
+
"learning_rate": 2.9473512783731334e-05,
|
| 1465 |
+
"loss": 1.8372,
|
| 1466 |
+
"step": 10400
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.8817821280904565,
|
| 1470 |
+
"grad_norm": 1.8369380235671997,
|
| 1471 |
+
"learning_rate": 2.947098135178466e-05,
|
| 1472 |
+
"loss": 1.8756,
|
| 1473 |
+
"step": 10450
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.8860011813349085,
|
| 1477 |
+
"grad_norm": 1.714089035987854,
|
| 1478 |
+
"learning_rate": 2.946844991983799e-05,
|
| 1479 |
+
"loss": 1.8421,
|
| 1480 |
+
"step": 10500
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.8902202345793604,
|
| 1484 |
+
"grad_norm": 2.0232396125793457,
|
| 1485 |
+
"learning_rate": 2.9465918487891318e-05,
|
| 1486 |
+
"loss": 1.8235,
|
| 1487 |
+
"step": 10550
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.8944392878238123,
|
| 1491 |
+
"grad_norm": 1.8890243768692017,
|
| 1492 |
+
"learning_rate": 2.9463387055944647e-05,
|
| 1493 |
+
"loss": 1.8769,
|
| 1494 |
+
"step": 10600
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.8986583410682643,
|
| 1498 |
+
"grad_norm": 1.7786074876785278,
|
| 1499 |
+
"learning_rate": 2.9460855623997976e-05,
|
| 1500 |
+
"loss": 1.8811,
|
| 1501 |
+
"step": 10650
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.9028773943127162,
|
| 1505 |
+
"grad_norm": 1.7798182964324951,
|
| 1506 |
+
"learning_rate": 2.9458324192051305e-05,
|
| 1507 |
+
"loss": 1.8511,
|
| 1508 |
+
"step": 10700
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.9070964475571681,
|
| 1512 |
+
"grad_norm": 2.3876659870147705,
|
| 1513 |
+
"learning_rate": 2.9455792760104634e-05,
|
| 1514 |
+
"loss": 1.8393,
|
| 1515 |
+
"step": 10750
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.9113155008016202,
|
| 1519 |
+
"grad_norm": 2.146610736846924,
|
| 1520 |
+
"learning_rate": 2.945326132815796e-05,
|
| 1521 |
+
"loss": 1.833,
|
| 1522 |
+
"step": 10800
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.9155345540460721,
|
| 1526 |
+
"grad_norm": 1.7398757934570312,
|
| 1527 |
+
"learning_rate": 2.9450729896211292e-05,
|
| 1528 |
+
"loss": 1.8093,
|
| 1529 |
+
"step": 10850
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.919753607290524,
|
| 1533 |
+
"grad_norm": 1.4934171438217163,
|
| 1534 |
+
"learning_rate": 2.944819846426462e-05,
|
| 1535 |
+
"loss": 1.8725,
|
| 1536 |
+
"step": 10900
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.923972660534976,
|
| 1540 |
+
"grad_norm": 1.6052733659744263,
|
| 1541 |
+
"learning_rate": 2.9445667032317947e-05,
|
| 1542 |
+
"loss": 1.8234,
|
| 1543 |
+
"step": 10950
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.9281917137794279,
|
| 1547 |
+
"grad_norm": 1.8470174074172974,
|
| 1548 |
+
"learning_rate": 2.9443135600371276e-05,
|
| 1549 |
+
"loss": 1.8085,
|
| 1550 |
+
"step": 11000
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.9324107670238798,
|
| 1554 |
+
"grad_norm": 1.607389211654663,
|
| 1555 |
+
"learning_rate": 2.9440604168424605e-05,
|
| 1556 |
+
"loss": 1.842,
|
| 1557 |
+
"step": 11050
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.9366298202683317,
|
| 1561 |
+
"grad_norm": 3.1832547187805176,
|
| 1562 |
+
"learning_rate": 2.9438072736477938e-05,
|
| 1563 |
+
"loss": 1.789,
|
| 1564 |
+
"step": 11100
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.9408488735127837,
|
| 1568 |
+
"grad_norm": 1.723089337348938,
|
| 1569 |
+
"learning_rate": 2.9435541304531264e-05,
|
| 1570 |
+
"loss": 1.8114,
|
| 1571 |
+
"step": 11150
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.9450679267572357,
|
| 1575 |
+
"grad_norm": 1.5779879093170166,
|
| 1576 |
+
"learning_rate": 2.9433009872584593e-05,
|
| 1577 |
+
"loss": 1.7647,
|
| 1578 |
+
"step": 11200
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.9492869800016877,
|
| 1582 |
+
"grad_norm": 1.6148508787155151,
|
| 1583 |
+
"learning_rate": 2.9430478440637922e-05,
|
| 1584 |
+
"loss": 1.8208,
|
| 1585 |
+
"step": 11250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.9535060332461396,
|
| 1589 |
+
"grad_norm": 1.9100016355514526,
|
| 1590 |
+
"learning_rate": 2.942794700869125e-05,
|
| 1591 |
+
"loss": 1.8349,
|
| 1592 |
+
"step": 11300
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.9577250864905915,
|
| 1596 |
+
"grad_norm": 1.6207829713821411,
|
| 1597 |
+
"learning_rate": 2.9425415576744577e-05,
|
| 1598 |
+
"loss": 1.8385,
|
| 1599 |
+
"step": 11350
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.9619441397350434,
|
| 1603 |
+
"grad_norm": 1.791765570640564,
|
| 1604 |
+
"learning_rate": 2.942288414479791e-05,
|
| 1605 |
+
"loss": 1.8213,
|
| 1606 |
+
"step": 11400
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.9661631929794954,
|
| 1610 |
+
"grad_norm": 1.7347838878631592,
|
| 1611 |
+
"learning_rate": 2.942035271285124e-05,
|
| 1612 |
+
"loss": 1.8133,
|
| 1613 |
+
"step": 11450
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.9703822462239473,
|
| 1617 |
+
"grad_norm": 1.8203942775726318,
|
| 1618 |
+
"learning_rate": 2.9417821280904564e-05,
|
| 1619 |
+
"loss": 1.7869,
|
| 1620 |
+
"step": 11500
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.9746012994683992,
|
| 1624 |
+
"grad_norm": 1.4434748888015747,
|
| 1625 |
+
"learning_rate": 2.9415289848957893e-05,
|
| 1626 |
+
"loss": 1.8252,
|
| 1627 |
+
"step": 11550
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.9788203527128513,
|
| 1631 |
+
"grad_norm": 1.7212822437286377,
|
| 1632 |
+
"learning_rate": 2.9412758417011222e-05,
|
| 1633 |
+
"loss": 1.8242,
|
| 1634 |
+
"step": 11600
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.9830394059573032,
|
| 1638 |
+
"grad_norm": 1.793928623199463,
|
| 1639 |
+
"learning_rate": 2.9410226985064555e-05,
|
| 1640 |
+
"loss": 1.8229,
|
| 1641 |
+
"step": 11650
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.9872584592017551,
|
| 1645 |
+
"grad_norm": 1.6043435335159302,
|
| 1646 |
+
"learning_rate": 2.940769555311788e-05,
|
| 1647 |
+
"loss": 1.8691,
|
| 1648 |
+
"step": 11700
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.9914775124462071,
|
| 1652 |
+
"grad_norm": 1.6892054080963135,
|
| 1653 |
+
"learning_rate": 2.940516412117121e-05,
|
| 1654 |
+
"loss": 1.797,
|
| 1655 |
+
"step": 11750
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.995696565690659,
|
| 1659 |
+
"grad_norm": 2.3594441413879395,
|
| 1660 |
+
"learning_rate": 2.940263268922454e-05,
|
| 1661 |
+
"loss": 1.8203,
|
| 1662 |
+
"step": 11800
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.9999156189351109,
|
| 1666 |
+
"grad_norm": 1.7604689598083496,
|
| 1667 |
+
"learning_rate": 2.9400101257277865e-05,
|
| 1668 |
+
"loss": 1.862,
|
| 1669 |
+
"step": 11850
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.0,
|
| 1673 |
+
"eval_loss": 1.6709089279174805,
|
| 1674 |
+
"eval_rouge1": 0.41913760479500367,
|
| 1675 |
+
"eval_rouge2": 0.20762530955850955,
|
| 1676 |
+
"eval_rougeL": 0.35250081833587826,
|
| 1677 |
+
"eval_runtime": 2942.3937,
|
| 1678 |
+
"eval_samples_per_second": 8.055,
|
| 1679 |
+
"eval_steps_per_second": 1.007,
|
| 1680 |
+
"step": 11851
|
| 1681 |
+
}
|
| 1682 |
+
],
|
| 1683 |
+
"logging_steps": 50,
|
| 1684 |
+
"max_steps": 592550,
|
| 1685 |
+
"num_input_tokens_seen": 0,
|
| 1686 |
+
"num_train_epochs": 50,
|
| 1687 |
+
"save_steps": 500,
|
| 1688 |
+
"stateful_callbacks": {
|
| 1689 |
+
"EarlyStoppingCallback": {
|
| 1690 |
+
"args": {
|
| 1691 |
+
"early_stopping_patience": 2,
|
| 1692 |
+
"early_stopping_threshold": 0.0
|
| 1693 |
+
},
|
| 1694 |
+
"attributes": {
|
| 1695 |
+
"early_stopping_patience_counter": 0
|
| 1696 |
+
}
|
| 1697 |
+
},
|
| 1698 |
+
"TrainerControl": {
|
| 1699 |
+
"args": {
|
| 1700 |
+
"should_epoch_stop": false,
|
| 1701 |
+
"should_evaluate": false,
|
| 1702 |
+
"should_log": false,
|
| 1703 |
+
"should_save": true,
|
| 1704 |
+
"should_training_stop": false
|
| 1705 |
+
},
|
| 1706 |
+
"attributes": {}
|
| 1707 |
+
}
|
| 1708 |
+
},
|
| 1709 |
+
"total_flos": 1.6965014011416576e+17,
|
| 1710 |
+
"train_batch_size": 8,
|
| 1711 |
+
"trial_name": null,
|
| 1712 |
+
"trial_params": null
|
| 1713 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5969
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1d2dc1a4c4abaafc5ec55711eddac0c51a74504ed2eb0c5750b0cd4ccf2723fe
|
| 3 |
size 5969
|