Upload 7 files
Browse files
checkpoint-23000/config.json
ADDED
@@ -0,0 +1,340 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract",
|
3 |
+
"architectures": [
|
4 |
+
"BertForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_dropout": null,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_dropout_prob": 0.1,
|
10 |
+
"hidden_size": 768,
|
11 |
+
"id2label": {
|
12 |
+
"0": "LABEL_0",
|
13 |
+
"1": "LABEL_1",
|
14 |
+
"2": "LABEL_2",
|
15 |
+
"3": "LABEL_3",
|
16 |
+
"4": "LABEL_4",
|
17 |
+
"5": "LABEL_5",
|
18 |
+
"6": "LABEL_6",
|
19 |
+
"7": "LABEL_7",
|
20 |
+
"8": "LABEL_8",
|
21 |
+
"9": "LABEL_9",
|
22 |
+
"10": "LABEL_10",
|
23 |
+
"11": "LABEL_11",
|
24 |
+
"12": "LABEL_12",
|
25 |
+
"13": "LABEL_13",
|
26 |
+
"14": "LABEL_14",
|
27 |
+
"15": "LABEL_15",
|
28 |
+
"16": "LABEL_16",
|
29 |
+
"17": "LABEL_17",
|
30 |
+
"18": "LABEL_18",
|
31 |
+
"19": "LABEL_19",
|
32 |
+
"20": "LABEL_20",
|
33 |
+
"21": "LABEL_21",
|
34 |
+
"22": "LABEL_22",
|
35 |
+
"23": "LABEL_23",
|
36 |
+
"24": "LABEL_24",
|
37 |
+
"25": "LABEL_25",
|
38 |
+
"26": "LABEL_26",
|
39 |
+
"27": "LABEL_27",
|
40 |
+
"28": "LABEL_28",
|
41 |
+
"29": "LABEL_29",
|
42 |
+
"30": "LABEL_30",
|
43 |
+
"31": "LABEL_31",
|
44 |
+
"32": "LABEL_32",
|
45 |
+
"33": "LABEL_33",
|
46 |
+
"34": "LABEL_34",
|
47 |
+
"35": "LABEL_35",
|
48 |
+
"36": "LABEL_36",
|
49 |
+
"37": "LABEL_37",
|
50 |
+
"38": "LABEL_38",
|
51 |
+
"39": "LABEL_39",
|
52 |
+
"40": "LABEL_40",
|
53 |
+
"41": "LABEL_41",
|
54 |
+
"42": "LABEL_42",
|
55 |
+
"43": "LABEL_43",
|
56 |
+
"44": "LABEL_44",
|
57 |
+
"45": "LABEL_45",
|
58 |
+
"46": "LABEL_46",
|
59 |
+
"47": "LABEL_47",
|
60 |
+
"48": "LABEL_48",
|
61 |
+
"49": "LABEL_49",
|
62 |
+
"50": "LABEL_50",
|
63 |
+
"51": "LABEL_51",
|
64 |
+
"52": "LABEL_52",
|
65 |
+
"53": "LABEL_53",
|
66 |
+
"54": "LABEL_54",
|
67 |
+
"55": "LABEL_55",
|
68 |
+
"56": "LABEL_56",
|
69 |
+
"57": "LABEL_57",
|
70 |
+
"58": "LABEL_58",
|
71 |
+
"59": "LABEL_59",
|
72 |
+
"60": "LABEL_60",
|
73 |
+
"61": "LABEL_61",
|
74 |
+
"62": "LABEL_62",
|
75 |
+
"63": "LABEL_63",
|
76 |
+
"64": "LABEL_64",
|
77 |
+
"65": "LABEL_65",
|
78 |
+
"66": "LABEL_66",
|
79 |
+
"67": "LABEL_67",
|
80 |
+
"68": "LABEL_68",
|
81 |
+
"69": "LABEL_69",
|
82 |
+
"70": "LABEL_70",
|
83 |
+
"71": "LABEL_71",
|
84 |
+
"72": "LABEL_72",
|
85 |
+
"73": "LABEL_73",
|
86 |
+
"74": "LABEL_74",
|
87 |
+
"75": "LABEL_75",
|
88 |
+
"76": "LABEL_76",
|
89 |
+
"77": "LABEL_77",
|
90 |
+
"78": "LABEL_78",
|
91 |
+
"79": "LABEL_79",
|
92 |
+
"80": "LABEL_80",
|
93 |
+
"81": "LABEL_81",
|
94 |
+
"82": "LABEL_82",
|
95 |
+
"83": "LABEL_83",
|
96 |
+
"84": "LABEL_84",
|
97 |
+
"85": "LABEL_85",
|
98 |
+
"86": "LABEL_86",
|
99 |
+
"87": "LABEL_87",
|
100 |
+
"88": "LABEL_88",
|
101 |
+
"89": "LABEL_89",
|
102 |
+
"90": "LABEL_90",
|
103 |
+
"91": "LABEL_91",
|
104 |
+
"92": "LABEL_92",
|
105 |
+
"93": "LABEL_93",
|
106 |
+
"94": "LABEL_94",
|
107 |
+
"95": "LABEL_95",
|
108 |
+
"96": "LABEL_96",
|
109 |
+
"97": "LABEL_97",
|
110 |
+
"98": "LABEL_98",
|
111 |
+
"99": "LABEL_99",
|
112 |
+
"100": "LABEL_100",
|
113 |
+
"101": "LABEL_101",
|
114 |
+
"102": "LABEL_102",
|
115 |
+
"103": "LABEL_103",
|
116 |
+
"104": "LABEL_104",
|
117 |
+
"105": "LABEL_105",
|
118 |
+
"106": "LABEL_106",
|
119 |
+
"107": "LABEL_107",
|
120 |
+
"108": "LABEL_108",
|
121 |
+
"109": "LABEL_109",
|
122 |
+
"110": "LABEL_110",
|
123 |
+
"111": "LABEL_111",
|
124 |
+
"112": "LABEL_112",
|
125 |
+
"113": "LABEL_113",
|
126 |
+
"114": "LABEL_114",
|
127 |
+
"115": "LABEL_115",
|
128 |
+
"116": "LABEL_116",
|
129 |
+
"117": "LABEL_117",
|
130 |
+
"118": "LABEL_118",
|
131 |
+
"119": "LABEL_119",
|
132 |
+
"120": "LABEL_120",
|
133 |
+
"121": "LABEL_121",
|
134 |
+
"122": "LABEL_122",
|
135 |
+
"123": "LABEL_123",
|
136 |
+
"124": "LABEL_124",
|
137 |
+
"125": "LABEL_125",
|
138 |
+
"126": "LABEL_126",
|
139 |
+
"127": "LABEL_127",
|
140 |
+
"128": "LABEL_128",
|
141 |
+
"129": "LABEL_129",
|
142 |
+
"130": "LABEL_130",
|
143 |
+
"131": "LABEL_131",
|
144 |
+
"132": "LABEL_132",
|
145 |
+
"133": "LABEL_133",
|
146 |
+
"134": "LABEL_134",
|
147 |
+
"135": "LABEL_135",
|
148 |
+
"136": "LABEL_136",
|
149 |
+
"137": "LABEL_137",
|
150 |
+
"138": "LABEL_138",
|
151 |
+
"139": "LABEL_139",
|
152 |
+
"140": "LABEL_140",
|
153 |
+
"141": "LABEL_141",
|
154 |
+
"142": "LABEL_142",
|
155 |
+
"143": "LABEL_143",
|
156 |
+
"144": "LABEL_144",
|
157 |
+
"145": "LABEL_145",
|
158 |
+
"146": "LABEL_146",
|
159 |
+
"147": "LABEL_147",
|
160 |
+
"148": "LABEL_148",
|
161 |
+
"149": "LABEL_149",
|
162 |
+
"150": "LABEL_150",
|
163 |
+
"151": "LABEL_151",
|
164 |
+
"152": "LABEL_152",
|
165 |
+
"153": "LABEL_153",
|
166 |
+
"154": "LABEL_154"
|
167 |
+
},
|
168 |
+
"initializer_range": 0.02,
|
169 |
+
"intermediate_size": 3072,
|
170 |
+
"label2id": {
|
171 |
+
"LABEL_0": 0,
|
172 |
+
"LABEL_1": 1,
|
173 |
+
"LABEL_10": 10,
|
174 |
+
"LABEL_100": 100,
|
175 |
+
"LABEL_101": 101,
|
176 |
+
"LABEL_102": 102,
|
177 |
+
"LABEL_103": 103,
|
178 |
+
"LABEL_104": 104,
|
179 |
+
"LABEL_105": 105,
|
180 |
+
"LABEL_106": 106,
|
181 |
+
"LABEL_107": 107,
|
182 |
+
"LABEL_108": 108,
|
183 |
+
"LABEL_109": 109,
|
184 |
+
"LABEL_11": 11,
|
185 |
+
"LABEL_110": 110,
|
186 |
+
"LABEL_111": 111,
|
187 |
+
"LABEL_112": 112,
|
188 |
+
"LABEL_113": 113,
|
189 |
+
"LABEL_114": 114,
|
190 |
+
"LABEL_115": 115,
|
191 |
+
"LABEL_116": 116,
|
192 |
+
"LABEL_117": 117,
|
193 |
+
"LABEL_118": 118,
|
194 |
+
"LABEL_119": 119,
|
195 |
+
"LABEL_12": 12,
|
196 |
+
"LABEL_120": 120,
|
197 |
+
"LABEL_121": 121,
|
198 |
+
"LABEL_122": 122,
|
199 |
+
"LABEL_123": 123,
|
200 |
+
"LABEL_124": 124,
|
201 |
+
"LABEL_125": 125,
|
202 |
+
"LABEL_126": 126,
|
203 |
+
"LABEL_127": 127,
|
204 |
+
"LABEL_128": 128,
|
205 |
+
"LABEL_129": 129,
|
206 |
+
"LABEL_13": 13,
|
207 |
+
"LABEL_130": 130,
|
208 |
+
"LABEL_131": 131,
|
209 |
+
"LABEL_132": 132,
|
210 |
+
"LABEL_133": 133,
|
211 |
+
"LABEL_134": 134,
|
212 |
+
"LABEL_135": 135,
|
213 |
+
"LABEL_136": 136,
|
214 |
+
"LABEL_137": 137,
|
215 |
+
"LABEL_138": 138,
|
216 |
+
"LABEL_139": 139,
|
217 |
+
"LABEL_14": 14,
|
218 |
+
"LABEL_140": 140,
|
219 |
+
"LABEL_141": 141,
|
220 |
+
"LABEL_142": 142,
|
221 |
+
"LABEL_143": 143,
|
222 |
+
"LABEL_144": 144,
|
223 |
+
"LABEL_145": 145,
|
224 |
+
"LABEL_146": 146,
|
225 |
+
"LABEL_147": 147,
|
226 |
+
"LABEL_148": 148,
|
227 |
+
"LABEL_149": 149,
|
228 |
+
"LABEL_15": 15,
|
229 |
+
"LABEL_150": 150,
|
230 |
+
"LABEL_151": 151,
|
231 |
+
"LABEL_152": 152,
|
232 |
+
"LABEL_153": 153,
|
233 |
+
"LABEL_154": 154,
|
234 |
+
"LABEL_16": 16,
|
235 |
+
"LABEL_17": 17,
|
236 |
+
"LABEL_18": 18,
|
237 |
+
"LABEL_19": 19,
|
238 |
+
"LABEL_2": 2,
|
239 |
+
"LABEL_20": 20,
|
240 |
+
"LABEL_21": 21,
|
241 |
+
"LABEL_22": 22,
|
242 |
+
"LABEL_23": 23,
|
243 |
+
"LABEL_24": 24,
|
244 |
+
"LABEL_25": 25,
|
245 |
+
"LABEL_26": 26,
|
246 |
+
"LABEL_27": 27,
|
247 |
+
"LABEL_28": 28,
|
248 |
+
"LABEL_29": 29,
|
249 |
+
"LABEL_3": 3,
|
250 |
+
"LABEL_30": 30,
|
251 |
+
"LABEL_31": 31,
|
252 |
+
"LABEL_32": 32,
|
253 |
+
"LABEL_33": 33,
|
254 |
+
"LABEL_34": 34,
|
255 |
+
"LABEL_35": 35,
|
256 |
+
"LABEL_36": 36,
|
257 |
+
"LABEL_37": 37,
|
258 |
+
"LABEL_38": 38,
|
259 |
+
"LABEL_39": 39,
|
260 |
+
"LABEL_4": 4,
|
261 |
+
"LABEL_40": 40,
|
262 |
+
"LABEL_41": 41,
|
263 |
+
"LABEL_42": 42,
|
264 |
+
"LABEL_43": 43,
|
265 |
+
"LABEL_44": 44,
|
266 |
+
"LABEL_45": 45,
|
267 |
+
"LABEL_46": 46,
|
268 |
+
"LABEL_47": 47,
|
269 |
+
"LABEL_48": 48,
|
270 |
+
"LABEL_49": 49,
|
271 |
+
"LABEL_5": 5,
|
272 |
+
"LABEL_50": 50,
|
273 |
+
"LABEL_51": 51,
|
274 |
+
"LABEL_52": 52,
|
275 |
+
"LABEL_53": 53,
|
276 |
+
"LABEL_54": 54,
|
277 |
+
"LABEL_55": 55,
|
278 |
+
"LABEL_56": 56,
|
279 |
+
"LABEL_57": 57,
|
280 |
+
"LABEL_58": 58,
|
281 |
+
"LABEL_59": 59,
|
282 |
+
"LABEL_6": 6,
|
283 |
+
"LABEL_60": 60,
|
284 |
+
"LABEL_61": 61,
|
285 |
+
"LABEL_62": 62,
|
286 |
+
"LABEL_63": 63,
|
287 |
+
"LABEL_64": 64,
|
288 |
+
"LABEL_65": 65,
|
289 |
+
"LABEL_66": 66,
|
290 |
+
"LABEL_67": 67,
|
291 |
+
"LABEL_68": 68,
|
292 |
+
"LABEL_69": 69,
|
293 |
+
"LABEL_7": 7,
|
294 |
+
"LABEL_70": 70,
|
295 |
+
"LABEL_71": 71,
|
296 |
+
"LABEL_72": 72,
|
297 |
+
"LABEL_73": 73,
|
298 |
+
"LABEL_74": 74,
|
299 |
+
"LABEL_75": 75,
|
300 |
+
"LABEL_76": 76,
|
301 |
+
"LABEL_77": 77,
|
302 |
+
"LABEL_78": 78,
|
303 |
+
"LABEL_79": 79,
|
304 |
+
"LABEL_8": 8,
|
305 |
+
"LABEL_80": 80,
|
306 |
+
"LABEL_81": 81,
|
307 |
+
"LABEL_82": 82,
|
308 |
+
"LABEL_83": 83,
|
309 |
+
"LABEL_84": 84,
|
310 |
+
"LABEL_85": 85,
|
311 |
+
"LABEL_86": 86,
|
312 |
+
"LABEL_87": 87,
|
313 |
+
"LABEL_88": 88,
|
314 |
+
"LABEL_89": 89,
|
315 |
+
"LABEL_9": 9,
|
316 |
+
"LABEL_90": 90,
|
317 |
+
"LABEL_91": 91,
|
318 |
+
"LABEL_92": 92,
|
319 |
+
"LABEL_93": 93,
|
320 |
+
"LABEL_94": 94,
|
321 |
+
"LABEL_95": 95,
|
322 |
+
"LABEL_96": 96,
|
323 |
+
"LABEL_97": 97,
|
324 |
+
"LABEL_98": 98,
|
325 |
+
"LABEL_99": 99
|
326 |
+
},
|
327 |
+
"layer_norm_eps": 1e-12,
|
328 |
+
"max_position_embeddings": 512,
|
329 |
+
"model_type": "bert",
|
330 |
+
"num_attention_heads": 12,
|
331 |
+
"num_hidden_layers": 12,
|
332 |
+
"pad_token_id": 0,
|
333 |
+
"position_embedding_type": "absolute",
|
334 |
+
"problem_type": "single_label_classification",
|
335 |
+
"torch_dtype": "float32",
|
336 |
+
"transformers_version": "4.49.0",
|
337 |
+
"type_vocab_size": 2,
|
338 |
+
"use_cache": true,
|
339 |
+
"vocab_size": 30522
|
340 |
+
}
|
checkpoint-23000/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7b51bc6475def37d86a8fa27d64b08bb41b95f6607e0f95e5761eea0096c84d5
|
3 |
+
size 438429276
|
checkpoint-23000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5edda25d345264622c24d1bb3e86446774a401b84ed848336d6eb7dc7d2f9a21
|
3 |
+
size 62397997
|
checkpoint-23000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2dae765b4076d6fd2c520098dca175703f65b5b3c120b93b109e6d08c0a39fb7
|
3 |
+
size 14575
|
checkpoint-23000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:eefefd20deb5318897955ad1f02eefc58ddd18d885dffa2b1f23f5f11a9d9579
|
3 |
+
size 627
|
checkpoint-23000/trainer_state.json
ADDED
@@ -0,0 +1,1643 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.23,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 23000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.001,
|
13 |
+
"grad_norm": 12.033066749572754,
|
14 |
+
"learning_rate": 9.998000000000002e-06,
|
15 |
+
"loss": 5.0328,
|
16 |
+
"step": 100
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.002,
|
20 |
+
"grad_norm": 12.140076637268066,
|
21 |
+
"learning_rate": 9.996e-06,
|
22 |
+
"loss": 4.8188,
|
23 |
+
"step": 200
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.003,
|
27 |
+
"grad_norm": 11.839982032775879,
|
28 |
+
"learning_rate": 9.994000000000001e-06,
|
29 |
+
"loss": 4.6021,
|
30 |
+
"step": 300
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.004,
|
34 |
+
"grad_norm": 13.72381591796875,
|
35 |
+
"learning_rate": 9.992e-06,
|
36 |
+
"loss": 4.4731,
|
37 |
+
"step": 400
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.005,
|
41 |
+
"grad_norm": 11.587430000305176,
|
42 |
+
"learning_rate": 9.990000000000001e-06,
|
43 |
+
"loss": 4.4251,
|
44 |
+
"step": 500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.006,
|
48 |
+
"grad_norm": 12.022308349609375,
|
49 |
+
"learning_rate": 9.988000000000002e-06,
|
50 |
+
"loss": 4.4416,
|
51 |
+
"step": 600
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.007,
|
55 |
+
"grad_norm": 12.038565635681152,
|
56 |
+
"learning_rate": 9.986e-06,
|
57 |
+
"loss": 4.3319,
|
58 |
+
"step": 700
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.008,
|
62 |
+
"grad_norm": 11.564003944396973,
|
63 |
+
"learning_rate": 9.984e-06,
|
64 |
+
"loss": 4.3267,
|
65 |
+
"step": 800
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.009,
|
69 |
+
"grad_norm": 11.518474578857422,
|
70 |
+
"learning_rate": 9.982e-06,
|
71 |
+
"loss": 4.2818,
|
72 |
+
"step": 900
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.01,
|
76 |
+
"grad_norm": 12.206191062927246,
|
77 |
+
"learning_rate": 9.980000000000001e-06,
|
78 |
+
"loss": 4.3167,
|
79 |
+
"step": 1000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.011,
|
83 |
+
"grad_norm": 11.940592765808105,
|
84 |
+
"learning_rate": 9.978000000000002e-06,
|
85 |
+
"loss": 4.3204,
|
86 |
+
"step": 1100
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.012,
|
90 |
+
"grad_norm": 12.109590530395508,
|
91 |
+
"learning_rate": 9.976e-06,
|
92 |
+
"loss": 4.3286,
|
93 |
+
"step": 1200
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.013,
|
97 |
+
"grad_norm": 12.218742370605469,
|
98 |
+
"learning_rate": 9.974e-06,
|
99 |
+
"loss": 4.4106,
|
100 |
+
"step": 1300
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.014,
|
104 |
+
"grad_norm": 11.53864860534668,
|
105 |
+
"learning_rate": 9.972e-06,
|
106 |
+
"loss": 4.3091,
|
107 |
+
"step": 1400
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.015,
|
111 |
+
"grad_norm": 13.800053596496582,
|
112 |
+
"learning_rate": 9.970000000000001e-06,
|
113 |
+
"loss": 4.2455,
|
114 |
+
"step": 1500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.016,
|
118 |
+
"grad_norm": 13.468424797058105,
|
119 |
+
"learning_rate": 9.968000000000002e-06,
|
120 |
+
"loss": 4.305,
|
121 |
+
"step": 1600
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.017,
|
125 |
+
"grad_norm": 11.736213684082031,
|
126 |
+
"learning_rate": 9.966e-06,
|
127 |
+
"loss": 4.3232,
|
128 |
+
"step": 1700
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.018,
|
132 |
+
"grad_norm": 10.893006324768066,
|
133 |
+
"learning_rate": 9.964e-06,
|
134 |
+
"loss": 4.3139,
|
135 |
+
"step": 1800
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.019,
|
139 |
+
"grad_norm": 14.407844543457031,
|
140 |
+
"learning_rate": 9.962e-06,
|
141 |
+
"loss": 4.2437,
|
142 |
+
"step": 1900
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.02,
|
146 |
+
"grad_norm": 11.889874458312988,
|
147 |
+
"learning_rate": 9.960000000000001e-06,
|
148 |
+
"loss": 4.2543,
|
149 |
+
"step": 2000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.021,
|
153 |
+
"grad_norm": 11.396763801574707,
|
154 |
+
"learning_rate": 9.958e-06,
|
155 |
+
"loss": 4.1477,
|
156 |
+
"step": 2100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.022,
|
160 |
+
"grad_norm": 13.99250316619873,
|
161 |
+
"learning_rate": 9.956000000000001e-06,
|
162 |
+
"loss": 4.248,
|
163 |
+
"step": 2200
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.023,
|
167 |
+
"grad_norm": 11.850446701049805,
|
168 |
+
"learning_rate": 9.954e-06,
|
169 |
+
"loss": 4.3014,
|
170 |
+
"step": 2300
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.024,
|
174 |
+
"grad_norm": 13.699837684631348,
|
175 |
+
"learning_rate": 9.952e-06,
|
176 |
+
"loss": 4.1937,
|
177 |
+
"step": 2400
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.025,
|
181 |
+
"grad_norm": 11.772696495056152,
|
182 |
+
"learning_rate": 9.950000000000001e-06,
|
183 |
+
"loss": 4.264,
|
184 |
+
"step": 2500
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.026,
|
188 |
+
"grad_norm": 13.460022926330566,
|
189 |
+
"learning_rate": 9.948e-06,
|
190 |
+
"loss": 4.2278,
|
191 |
+
"step": 2600
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.027,
|
195 |
+
"grad_norm": 11.80987548828125,
|
196 |
+
"learning_rate": 9.946000000000001e-06,
|
197 |
+
"loss": 4.2416,
|
198 |
+
"step": 2700
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.028,
|
202 |
+
"grad_norm": 12.391595840454102,
|
203 |
+
"learning_rate": 9.944e-06,
|
204 |
+
"loss": 4.2183,
|
205 |
+
"step": 2800
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.029,
|
209 |
+
"grad_norm": 12.336369514465332,
|
210 |
+
"learning_rate": 9.942e-06,
|
211 |
+
"loss": 4.225,
|
212 |
+
"step": 2900
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.03,
|
216 |
+
"grad_norm": 12.137269020080566,
|
217 |
+
"learning_rate": 9.940000000000001e-06,
|
218 |
+
"loss": 4.2045,
|
219 |
+
"step": 3000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.031,
|
223 |
+
"grad_norm": 12.397940635681152,
|
224 |
+
"learning_rate": 9.938e-06,
|
225 |
+
"loss": 4.1523,
|
226 |
+
"step": 3100
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.032,
|
230 |
+
"grad_norm": 12.940911293029785,
|
231 |
+
"learning_rate": 9.936000000000001e-06,
|
232 |
+
"loss": 4.0193,
|
233 |
+
"step": 3200
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.033,
|
237 |
+
"grad_norm": 16.68646812438965,
|
238 |
+
"learning_rate": 9.934e-06,
|
239 |
+
"loss": 4.1939,
|
240 |
+
"step": 3300
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.034,
|
244 |
+
"grad_norm": 12.541526794433594,
|
245 |
+
"learning_rate": 9.932e-06,
|
246 |
+
"loss": 4.0829,
|
247 |
+
"step": 3400
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.035,
|
251 |
+
"grad_norm": 11.975446701049805,
|
252 |
+
"learning_rate": 9.930000000000001e-06,
|
253 |
+
"loss": 4.2043,
|
254 |
+
"step": 3500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.036,
|
258 |
+
"grad_norm": 12.638479232788086,
|
259 |
+
"learning_rate": 9.928e-06,
|
260 |
+
"loss": 4.1315,
|
261 |
+
"step": 3600
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.037,
|
265 |
+
"grad_norm": 13.302968978881836,
|
266 |
+
"learning_rate": 9.926000000000001e-06,
|
267 |
+
"loss": 4.0406,
|
268 |
+
"step": 3700
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.038,
|
272 |
+
"grad_norm": 12.131388664245605,
|
273 |
+
"learning_rate": 9.924e-06,
|
274 |
+
"loss": 4.0571,
|
275 |
+
"step": 3800
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.039,
|
279 |
+
"grad_norm": 13.895309448242188,
|
280 |
+
"learning_rate": 9.922000000000001e-06,
|
281 |
+
"loss": 4.0597,
|
282 |
+
"step": 3900
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.04,
|
286 |
+
"grad_norm": 15.263091087341309,
|
287 |
+
"learning_rate": 9.920000000000002e-06,
|
288 |
+
"loss": 4.1587,
|
289 |
+
"step": 4000
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.041,
|
293 |
+
"grad_norm": 12.314478874206543,
|
294 |
+
"learning_rate": 9.918e-06,
|
295 |
+
"loss": 4.0695,
|
296 |
+
"step": 4100
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.042,
|
300 |
+
"grad_norm": 13.542490005493164,
|
301 |
+
"learning_rate": 9.916000000000001e-06,
|
302 |
+
"loss": 4.088,
|
303 |
+
"step": 4200
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.043,
|
307 |
+
"grad_norm": 14.835192680358887,
|
308 |
+
"learning_rate": 9.914e-06,
|
309 |
+
"loss": 4.1067,
|
310 |
+
"step": 4300
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.044,
|
314 |
+
"grad_norm": 13.269238471984863,
|
315 |
+
"learning_rate": 9.912000000000001e-06,
|
316 |
+
"loss": 4.0273,
|
317 |
+
"step": 4400
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.045,
|
321 |
+
"grad_norm": 12.532042503356934,
|
322 |
+
"learning_rate": 9.91e-06,
|
323 |
+
"loss": 3.9738,
|
324 |
+
"step": 4500
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.046,
|
328 |
+
"grad_norm": 14.506613731384277,
|
329 |
+
"learning_rate": 9.908e-06,
|
330 |
+
"loss": 4.0215,
|
331 |
+
"step": 4600
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.047,
|
335 |
+
"grad_norm": 12.91763973236084,
|
336 |
+
"learning_rate": 9.906000000000001e-06,
|
337 |
+
"loss": 3.9976,
|
338 |
+
"step": 4700
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.048,
|
342 |
+
"grad_norm": 12.20261001586914,
|
343 |
+
"learning_rate": 9.904e-06,
|
344 |
+
"loss": 3.9172,
|
345 |
+
"step": 4800
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.049,
|
349 |
+
"grad_norm": 13.156211853027344,
|
350 |
+
"learning_rate": 9.902000000000001e-06,
|
351 |
+
"loss": 3.9871,
|
352 |
+
"step": 4900
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.05,
|
356 |
+
"grad_norm": 13.59281063079834,
|
357 |
+
"learning_rate": 9.9e-06,
|
358 |
+
"loss": 3.9724,
|
359 |
+
"step": 5000
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.051,
|
363 |
+
"grad_norm": 13.202598571777344,
|
364 |
+
"learning_rate": 9.898e-06,
|
365 |
+
"loss": 3.8869,
|
366 |
+
"step": 5100
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.052,
|
370 |
+
"grad_norm": 16.82631492614746,
|
371 |
+
"learning_rate": 9.896000000000001e-06,
|
372 |
+
"loss": 4.0354,
|
373 |
+
"step": 5200
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.053,
|
377 |
+
"grad_norm": 14.205794334411621,
|
378 |
+
"learning_rate": 9.894e-06,
|
379 |
+
"loss": 4.0322,
|
380 |
+
"step": 5300
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.054,
|
384 |
+
"grad_norm": 14.135138511657715,
|
385 |
+
"learning_rate": 9.892000000000001e-06,
|
386 |
+
"loss": 4.01,
|
387 |
+
"step": 5400
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.055,
|
391 |
+
"grad_norm": 14.365863800048828,
|
392 |
+
"learning_rate": 9.89e-06,
|
393 |
+
"loss": 3.9914,
|
394 |
+
"step": 5500
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.056,
|
398 |
+
"grad_norm": 14.389483451843262,
|
399 |
+
"learning_rate": 9.888000000000001e-06,
|
400 |
+
"loss": 3.8672,
|
401 |
+
"step": 5600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.057,
|
405 |
+
"grad_norm": 12.534111976623535,
|
406 |
+
"learning_rate": 9.886000000000002e-06,
|
407 |
+
"loss": 3.8888,
|
408 |
+
"step": 5700
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.058,
|
412 |
+
"grad_norm": 12.352408409118652,
|
413 |
+
"learning_rate": 9.884e-06,
|
414 |
+
"loss": 3.9466,
|
415 |
+
"step": 5800
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.059,
|
419 |
+
"grad_norm": 11.810559272766113,
|
420 |
+
"learning_rate": 9.882000000000001e-06,
|
421 |
+
"loss": 3.8322,
|
422 |
+
"step": 5900
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.06,
|
426 |
+
"grad_norm": 15.00880241394043,
|
427 |
+
"learning_rate": 9.88e-06,
|
428 |
+
"loss": 3.9209,
|
429 |
+
"step": 6000
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.061,
|
433 |
+
"grad_norm": 13.07725715637207,
|
434 |
+
"learning_rate": 9.878000000000001e-06,
|
435 |
+
"loss": 3.935,
|
436 |
+
"step": 6100
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.062,
|
440 |
+
"grad_norm": 11.981454849243164,
|
441 |
+
"learning_rate": 9.876000000000002e-06,
|
442 |
+
"loss": 3.808,
|
443 |
+
"step": 6200
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.063,
|
447 |
+
"grad_norm": 14.732353210449219,
|
448 |
+
"learning_rate": 9.874e-06,
|
449 |
+
"loss": 3.9113,
|
450 |
+
"step": 6300
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.064,
|
454 |
+
"grad_norm": 14.65374755859375,
|
455 |
+
"learning_rate": 9.872e-06,
|
456 |
+
"loss": 3.9368,
|
457 |
+
"step": 6400
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.065,
|
461 |
+
"grad_norm": 19.678359985351562,
|
462 |
+
"learning_rate": 9.87e-06,
|
463 |
+
"loss": 3.8806,
|
464 |
+
"step": 6500
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.066,
|
468 |
+
"grad_norm": 13.10519790649414,
|
469 |
+
"learning_rate": 9.868000000000001e-06,
|
470 |
+
"loss": 3.7696,
|
471 |
+
"step": 6600
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.067,
|
475 |
+
"grad_norm": 13.150310516357422,
|
476 |
+
"learning_rate": 9.866000000000002e-06,
|
477 |
+
"loss": 3.7716,
|
478 |
+
"step": 6700
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.068,
|
482 |
+
"grad_norm": 12.890090942382812,
|
483 |
+
"learning_rate": 9.864e-06,
|
484 |
+
"loss": 3.8475,
|
485 |
+
"step": 6800
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.069,
|
489 |
+
"grad_norm": 14.92232894897461,
|
490 |
+
"learning_rate": 9.862e-06,
|
491 |
+
"loss": 3.9018,
|
492 |
+
"step": 6900
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.07,
|
496 |
+
"grad_norm": 17.389636993408203,
|
497 |
+
"learning_rate": 9.86e-06,
|
498 |
+
"loss": 3.8824,
|
499 |
+
"step": 7000
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.071,
|
503 |
+
"grad_norm": 13.9534330368042,
|
504 |
+
"learning_rate": 9.858000000000001e-06,
|
505 |
+
"loss": 3.8621,
|
506 |
+
"step": 7100
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.072,
|
510 |
+
"grad_norm": 13.286906242370605,
|
511 |
+
"learning_rate": 9.856000000000002e-06,
|
512 |
+
"loss": 3.9293,
|
513 |
+
"step": 7200
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.073,
|
517 |
+
"grad_norm": 13.590744972229004,
|
518 |
+
"learning_rate": 9.854000000000001e-06,
|
519 |
+
"loss": 3.8883,
|
520 |
+
"step": 7300
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.074,
|
524 |
+
"grad_norm": 10.841891288757324,
|
525 |
+
"learning_rate": 9.852e-06,
|
526 |
+
"loss": 3.8022,
|
527 |
+
"step": 7400
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.075,
|
531 |
+
"grad_norm": 14.516188621520996,
|
532 |
+
"learning_rate": 9.85e-06,
|
533 |
+
"loss": 3.713,
|
534 |
+
"step": 7500
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.076,
|
538 |
+
"grad_norm": 16.878511428833008,
|
539 |
+
"learning_rate": 9.848000000000001e-06,
|
540 |
+
"loss": 3.8758,
|
541 |
+
"step": 7600
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.077,
|
545 |
+
"grad_norm": 16.681041717529297,
|
546 |
+
"learning_rate": 9.846000000000002e-06,
|
547 |
+
"loss": 3.7602,
|
548 |
+
"step": 7700
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.078,
|
552 |
+
"grad_norm": 14.792035102844238,
|
553 |
+
"learning_rate": 9.844000000000001e-06,
|
554 |
+
"loss": 3.7145,
|
555 |
+
"step": 7800
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.079,
|
559 |
+
"grad_norm": 15.74644660949707,
|
560 |
+
"learning_rate": 9.842e-06,
|
561 |
+
"loss": 3.7551,
|
562 |
+
"step": 7900
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.08,
|
566 |
+
"grad_norm": 16.060367584228516,
|
567 |
+
"learning_rate": 9.84e-06,
|
568 |
+
"loss": 3.7701,
|
569 |
+
"step": 8000
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.081,
|
573 |
+
"grad_norm": 13.487128257751465,
|
574 |
+
"learning_rate": 9.838000000000001e-06,
|
575 |
+
"loss": 3.6881,
|
576 |
+
"step": 8100
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.082,
|
580 |
+
"grad_norm": 15.940110206604004,
|
581 |
+
"learning_rate": 9.836e-06,
|
582 |
+
"loss": 3.9418,
|
583 |
+
"step": 8200
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.083,
|
587 |
+
"grad_norm": 13.603132247924805,
|
588 |
+
"learning_rate": 9.834000000000001e-06,
|
589 |
+
"loss": 3.7518,
|
590 |
+
"step": 8300
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.084,
|
594 |
+
"grad_norm": 17.045244216918945,
|
595 |
+
"learning_rate": 9.832e-06,
|
596 |
+
"loss": 3.7167,
|
597 |
+
"step": 8400
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.085,
|
601 |
+
"grad_norm": 12.39263916015625,
|
602 |
+
"learning_rate": 9.83e-06,
|
603 |
+
"loss": 3.6829,
|
604 |
+
"step": 8500
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.086,
|
608 |
+
"grad_norm": 12.677363395690918,
|
609 |
+
"learning_rate": 9.828000000000001e-06,
|
610 |
+
"loss": 3.7458,
|
611 |
+
"step": 8600
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.087,
|
615 |
+
"grad_norm": 15.024678230285645,
|
616 |
+
"learning_rate": 9.826e-06,
|
617 |
+
"loss": 3.7334,
|
618 |
+
"step": 8700
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.088,
|
622 |
+
"grad_norm": 17.41254997253418,
|
623 |
+
"learning_rate": 9.824000000000001e-06,
|
624 |
+
"loss": 3.7704,
|
625 |
+
"step": 8800
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.089,
|
629 |
+
"grad_norm": 19.782014846801758,
|
630 |
+
"learning_rate": 9.822e-06,
|
631 |
+
"loss": 3.6834,
|
632 |
+
"step": 8900
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.09,
|
636 |
+
"grad_norm": 16.899019241333008,
|
637 |
+
"learning_rate": 9.820000000000001e-06,
|
638 |
+
"loss": 3.7304,
|
639 |
+
"step": 9000
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.091,
|
643 |
+
"grad_norm": 14.481075286865234,
|
644 |
+
"learning_rate": 9.818000000000002e-06,
|
645 |
+
"loss": 3.7307,
|
646 |
+
"step": 9100
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.092,
|
650 |
+
"grad_norm": 18.121864318847656,
|
651 |
+
"learning_rate": 9.816e-06,
|
652 |
+
"loss": 3.6793,
|
653 |
+
"step": 9200
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.093,
|
657 |
+
"grad_norm": 15.916873931884766,
|
658 |
+
"learning_rate": 9.814000000000001e-06,
|
659 |
+
"loss": 3.7664,
|
660 |
+
"step": 9300
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.094,
|
664 |
+
"grad_norm": 18.305234909057617,
|
665 |
+
"learning_rate": 9.812e-06,
|
666 |
+
"loss": 3.6887,
|
667 |
+
"step": 9400
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.095,
|
671 |
+
"grad_norm": 18.262725830078125,
|
672 |
+
"learning_rate": 9.810000000000001e-06,
|
673 |
+
"loss": 3.6865,
|
674 |
+
"step": 9500
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.096,
|
678 |
+
"grad_norm": 21.94981575012207,
|
679 |
+
"learning_rate": 9.808000000000002e-06,
|
680 |
+
"loss": 3.5916,
|
681 |
+
"step": 9600
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.097,
|
685 |
+
"grad_norm": 15.031508445739746,
|
686 |
+
"learning_rate": 9.806e-06,
|
687 |
+
"loss": 3.5732,
|
688 |
+
"step": 9700
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.098,
|
692 |
+
"grad_norm": 13.64002799987793,
|
693 |
+
"learning_rate": 9.804000000000001e-06,
|
694 |
+
"loss": 3.6104,
|
695 |
+
"step": 9800
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.099,
|
699 |
+
"grad_norm": 22.877960205078125,
|
700 |
+
"learning_rate": 9.802e-06,
|
701 |
+
"loss": 3.622,
|
702 |
+
"step": 9900
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.1,
|
706 |
+
"grad_norm": 13.404180526733398,
|
707 |
+
"learning_rate": 9.800000000000001e-06,
|
708 |
+
"loss": 3.6693,
|
709 |
+
"step": 10000
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.101,
|
713 |
+
"grad_norm": 14.348539352416992,
|
714 |
+
"learning_rate": 9.798e-06,
|
715 |
+
"loss": 3.4753,
|
716 |
+
"step": 10100
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.102,
|
720 |
+
"grad_norm": 15.996590614318848,
|
721 |
+
"learning_rate": 9.796e-06,
|
722 |
+
"loss": 3.5059,
|
723 |
+
"step": 10200
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.103,
|
727 |
+
"grad_norm": 16.6004638671875,
|
728 |
+
"learning_rate": 9.794000000000001e-06,
|
729 |
+
"loss": 3.7168,
|
730 |
+
"step": 10300
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.104,
|
734 |
+
"grad_norm": 22.660940170288086,
|
735 |
+
"learning_rate": 9.792e-06,
|
736 |
+
"loss": 3.5521,
|
737 |
+
"step": 10400
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.105,
|
741 |
+
"grad_norm": 16.634521484375,
|
742 |
+
"learning_rate": 9.790000000000001e-06,
|
743 |
+
"loss": 3.5894,
|
744 |
+
"step": 10500
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.106,
|
748 |
+
"grad_norm": 17.89203643798828,
|
749 |
+
"learning_rate": 9.788e-06,
|
750 |
+
"loss": 3.5524,
|
751 |
+
"step": 10600
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.107,
|
755 |
+
"grad_norm": 17.027833938598633,
|
756 |
+
"learning_rate": 9.786e-06,
|
757 |
+
"loss": 3.5953,
|
758 |
+
"step": 10700
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.108,
|
762 |
+
"grad_norm": 19.78436279296875,
|
763 |
+
"learning_rate": 9.784000000000002e-06,
|
764 |
+
"loss": 3.6688,
|
765 |
+
"step": 10800
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.109,
|
769 |
+
"grad_norm": 17.20643424987793,
|
770 |
+
"learning_rate": 9.782e-06,
|
771 |
+
"loss": 3.5023,
|
772 |
+
"step": 10900
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.11,
|
776 |
+
"grad_norm": 14.809402465820312,
|
777 |
+
"learning_rate": 9.780000000000001e-06,
|
778 |
+
"loss": 3.4398,
|
779 |
+
"step": 11000
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.111,
|
783 |
+
"grad_norm": 21.907175064086914,
|
784 |
+
"learning_rate": 9.778e-06,
|
785 |
+
"loss": 3.5254,
|
786 |
+
"step": 11100
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.112,
|
790 |
+
"grad_norm": 23.719179153442383,
|
791 |
+
"learning_rate": 9.776000000000001e-06,
|
792 |
+
"loss": 3.4801,
|
793 |
+
"step": 11200
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.113,
|
797 |
+
"grad_norm": 15.437023162841797,
|
798 |
+
"learning_rate": 9.774000000000002e-06,
|
799 |
+
"loss": 3.5724,
|
800 |
+
"step": 11300
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.114,
|
804 |
+
"grad_norm": 21.82857322692871,
|
805 |
+
"learning_rate": 9.772e-06,
|
806 |
+
"loss": 3.5787,
|
807 |
+
"step": 11400
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.115,
|
811 |
+
"grad_norm": 20.26848602294922,
|
812 |
+
"learning_rate": 9.770000000000001e-06,
|
813 |
+
"loss": 3.6333,
|
814 |
+
"step": 11500
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.116,
|
818 |
+
"grad_norm": 17.0762882232666,
|
819 |
+
"learning_rate": 9.768e-06,
|
820 |
+
"loss": 3.3418,
|
821 |
+
"step": 11600
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.117,
|
825 |
+
"grad_norm": 20.440383911132812,
|
826 |
+
"learning_rate": 9.766000000000001e-06,
|
827 |
+
"loss": 3.5072,
|
828 |
+
"step": 11700
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.118,
|
832 |
+
"grad_norm": 17.19301986694336,
|
833 |
+
"learning_rate": 9.764000000000002e-06,
|
834 |
+
"loss": 3.491,
|
835 |
+
"step": 11800
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.119,
|
839 |
+
"grad_norm": 20.88847541809082,
|
840 |
+
"learning_rate": 9.762e-06,
|
841 |
+
"loss": 3.5482,
|
842 |
+
"step": 11900
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.12,
|
846 |
+
"grad_norm": 17.677921295166016,
|
847 |
+
"learning_rate": 9.760000000000001e-06,
|
848 |
+
"loss": 3.5495,
|
849 |
+
"step": 12000
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.121,
|
853 |
+
"grad_norm": 19.91204833984375,
|
854 |
+
"learning_rate": 9.758e-06,
|
855 |
+
"loss": 3.652,
|
856 |
+
"step": 12100
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.122,
|
860 |
+
"grad_norm": 39.53171920776367,
|
861 |
+
"learning_rate": 9.756000000000001e-06,
|
862 |
+
"loss": 3.5966,
|
863 |
+
"step": 12200
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.123,
|
867 |
+
"grad_norm": 15.958291053771973,
|
868 |
+
"learning_rate": 9.754000000000002e-06,
|
869 |
+
"loss": 3.5364,
|
870 |
+
"step": 12300
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.124,
|
874 |
+
"grad_norm": 15.132060050964355,
|
875 |
+
"learning_rate": 9.752e-06,
|
876 |
+
"loss": 3.468,
|
877 |
+
"step": 12400
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.125,
|
881 |
+
"grad_norm": 14.320738792419434,
|
882 |
+
"learning_rate": 9.75e-06,
|
883 |
+
"loss": 3.3429,
|
884 |
+
"step": 12500
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.126,
|
888 |
+
"grad_norm": 20.6189022064209,
|
889 |
+
"learning_rate": 9.748e-06,
|
890 |
+
"loss": 3.3507,
|
891 |
+
"step": 12600
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.127,
|
895 |
+
"grad_norm": 14.898783683776855,
|
896 |
+
"learning_rate": 9.746000000000001e-06,
|
897 |
+
"loss": 3.3785,
|
898 |
+
"step": 12700
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.128,
|
902 |
+
"grad_norm": 22.72285270690918,
|
903 |
+
"learning_rate": 9.744000000000002e-06,
|
904 |
+
"loss": 3.5126,
|
905 |
+
"step": 12800
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.129,
|
909 |
+
"grad_norm": 22.329116821289062,
|
910 |
+
"learning_rate": 9.742000000000001e-06,
|
911 |
+
"loss": 3.4503,
|
912 |
+
"step": 12900
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.13,
|
916 |
+
"grad_norm": 18.049467086791992,
|
917 |
+
"learning_rate": 9.74e-06,
|
918 |
+
"loss": 3.4891,
|
919 |
+
"step": 13000
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.131,
|
923 |
+
"grad_norm": 14.28784465789795,
|
924 |
+
"learning_rate": 9.738e-06,
|
925 |
+
"loss": 3.33,
|
926 |
+
"step": 13100
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.132,
|
930 |
+
"grad_norm": 19.659822463989258,
|
931 |
+
"learning_rate": 9.736000000000001e-06,
|
932 |
+
"loss": 3.4724,
|
933 |
+
"step": 13200
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.133,
|
937 |
+
"grad_norm": 19.972923278808594,
|
938 |
+
"learning_rate": 9.734000000000002e-06,
|
939 |
+
"loss": 3.3965,
|
940 |
+
"step": 13300
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.134,
|
944 |
+
"grad_norm": 21.733108520507812,
|
945 |
+
"learning_rate": 9.732000000000001e-06,
|
946 |
+
"loss": 3.584,
|
947 |
+
"step": 13400
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.135,
|
951 |
+
"grad_norm": 13.769856452941895,
|
952 |
+
"learning_rate": 9.73e-06,
|
953 |
+
"loss": 3.4789,
|
954 |
+
"step": 13500
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.136,
|
958 |
+
"grad_norm": 15.672243118286133,
|
959 |
+
"learning_rate": 9.728e-06,
|
960 |
+
"loss": 3.5081,
|
961 |
+
"step": 13600
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.137,
|
965 |
+
"grad_norm": 17.671894073486328,
|
966 |
+
"learning_rate": 9.726000000000001e-06,
|
967 |
+
"loss": 3.3696,
|
968 |
+
"step": 13700
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.138,
|
972 |
+
"grad_norm": 19.69550323486328,
|
973 |
+
"learning_rate": 9.724e-06,
|
974 |
+
"loss": 3.3707,
|
975 |
+
"step": 13800
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.139,
|
979 |
+
"grad_norm": 14.621719360351562,
|
980 |
+
"learning_rate": 9.722000000000001e-06,
|
981 |
+
"loss": 3.4709,
|
982 |
+
"step": 13900
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.14,
|
986 |
+
"grad_norm": 17.52949333190918,
|
987 |
+
"learning_rate": 9.72e-06,
|
988 |
+
"loss": 3.4979,
|
989 |
+
"step": 14000
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.141,
|
993 |
+
"grad_norm": 15.679729461669922,
|
994 |
+
"learning_rate": 9.718e-06,
|
995 |
+
"loss": 3.4584,
|
996 |
+
"step": 14100
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.142,
|
1000 |
+
"grad_norm": 17.527435302734375,
|
1001 |
+
"learning_rate": 9.716000000000002e-06,
|
1002 |
+
"loss": 3.3812,
|
1003 |
+
"step": 14200
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.143,
|
1007 |
+
"grad_norm": 24.084278106689453,
|
1008 |
+
"learning_rate": 9.714e-06,
|
1009 |
+
"loss": 3.4052,
|
1010 |
+
"step": 14300
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.144,
|
1014 |
+
"grad_norm": 20.039127349853516,
|
1015 |
+
"learning_rate": 9.712e-06,
|
1016 |
+
"loss": 3.3659,
|
1017 |
+
"step": 14400
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.145,
|
1021 |
+
"grad_norm": 18.518939971923828,
|
1022 |
+
"learning_rate": 9.71e-06,
|
1023 |
+
"loss": 3.4249,
|
1024 |
+
"step": 14500
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.146,
|
1028 |
+
"grad_norm": 19.596946716308594,
|
1029 |
+
"learning_rate": 9.708000000000001e-06,
|
1030 |
+
"loss": 3.3909,
|
1031 |
+
"step": 14600
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.147,
|
1035 |
+
"grad_norm": 14.774336814880371,
|
1036 |
+
"learning_rate": 9.706000000000002e-06,
|
1037 |
+
"loss": 3.3216,
|
1038 |
+
"step": 14700
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.148,
|
1042 |
+
"grad_norm": 27.980627059936523,
|
1043 |
+
"learning_rate": 9.704e-06,
|
1044 |
+
"loss": 3.3794,
|
1045 |
+
"step": 14800
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.149,
|
1049 |
+
"grad_norm": 16.481491088867188,
|
1050 |
+
"learning_rate": 9.702e-06,
|
1051 |
+
"loss": 3.4138,
|
1052 |
+
"step": 14900
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.15,
|
1056 |
+
"grad_norm": 18.48386573791504,
|
1057 |
+
"learning_rate": 9.7e-06,
|
1058 |
+
"loss": 3.3635,
|
1059 |
+
"step": 15000
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.151,
|
1063 |
+
"grad_norm": 14.089752197265625,
|
1064 |
+
"learning_rate": 9.698000000000001e-06,
|
1065 |
+
"loss": 3.363,
|
1066 |
+
"step": 15100
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.152,
|
1070 |
+
"grad_norm": 15.205988883972168,
|
1071 |
+
"learning_rate": 9.696000000000002e-06,
|
1072 |
+
"loss": 3.3038,
|
1073 |
+
"step": 15200
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.153,
|
1077 |
+
"grad_norm": 18.17800521850586,
|
1078 |
+
"learning_rate": 9.694e-06,
|
1079 |
+
"loss": 3.2748,
|
1080 |
+
"step": 15300
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.154,
|
1084 |
+
"grad_norm": 15.958276748657227,
|
1085 |
+
"learning_rate": 9.692e-06,
|
1086 |
+
"loss": 3.2809,
|
1087 |
+
"step": 15400
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.155,
|
1091 |
+
"grad_norm": 20.20997428894043,
|
1092 |
+
"learning_rate": 9.69e-06,
|
1093 |
+
"loss": 3.3366,
|
1094 |
+
"step": 15500
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.156,
|
1098 |
+
"grad_norm": 13.844518661499023,
|
1099 |
+
"learning_rate": 9.688000000000001e-06,
|
1100 |
+
"loss": 3.424,
|
1101 |
+
"step": 15600
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.157,
|
1105 |
+
"grad_norm": 28.56269645690918,
|
1106 |
+
"learning_rate": 9.686000000000002e-06,
|
1107 |
+
"loss": 3.4567,
|
1108 |
+
"step": 15700
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.158,
|
1112 |
+
"grad_norm": 14.436336517333984,
|
1113 |
+
"learning_rate": 9.684e-06,
|
1114 |
+
"loss": 3.2089,
|
1115 |
+
"step": 15800
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.159,
|
1119 |
+
"grad_norm": 28.26078987121582,
|
1120 |
+
"learning_rate": 9.682e-06,
|
1121 |
+
"loss": 3.439,
|
1122 |
+
"step": 15900
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.16,
|
1126 |
+
"grad_norm": 19.394569396972656,
|
1127 |
+
"learning_rate": 9.68e-06,
|
1128 |
+
"loss": 3.3376,
|
1129 |
+
"step": 16000
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.161,
|
1133 |
+
"grad_norm": 18.12739372253418,
|
1134 |
+
"learning_rate": 9.678000000000001e-06,
|
1135 |
+
"loss": 3.2295,
|
1136 |
+
"step": 16100
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.162,
|
1140 |
+
"grad_norm": 20.420162200927734,
|
1141 |
+
"learning_rate": 9.676e-06,
|
1142 |
+
"loss": 3.307,
|
1143 |
+
"step": 16200
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.163,
|
1147 |
+
"grad_norm": 15.34536361694336,
|
1148 |
+
"learning_rate": 9.674000000000001e-06,
|
1149 |
+
"loss": 3.3014,
|
1150 |
+
"step": 16300
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.164,
|
1154 |
+
"grad_norm": 20.239303588867188,
|
1155 |
+
"learning_rate": 9.672e-06,
|
1156 |
+
"loss": 3.3339,
|
1157 |
+
"step": 16400
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.165,
|
1161 |
+
"grad_norm": 15.123329162597656,
|
1162 |
+
"learning_rate": 9.67e-06,
|
1163 |
+
"loss": 3.3073,
|
1164 |
+
"step": 16500
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.166,
|
1168 |
+
"grad_norm": 21.282299041748047,
|
1169 |
+
"learning_rate": 9.668000000000001e-06,
|
1170 |
+
"loss": 3.2809,
|
1171 |
+
"step": 16600
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.167,
|
1175 |
+
"grad_norm": 18.039140701293945,
|
1176 |
+
"learning_rate": 9.666e-06,
|
1177 |
+
"loss": 3.3368,
|
1178 |
+
"step": 16700
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.168,
|
1182 |
+
"grad_norm": 17.391878128051758,
|
1183 |
+
"learning_rate": 9.664000000000001e-06,
|
1184 |
+
"loss": 3.2135,
|
1185 |
+
"step": 16800
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.169,
|
1189 |
+
"grad_norm": 29.216005325317383,
|
1190 |
+
"learning_rate": 9.662e-06,
|
1191 |
+
"loss": 3.2238,
|
1192 |
+
"step": 16900
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 0.17,
|
1196 |
+
"grad_norm": 24.716182708740234,
|
1197 |
+
"learning_rate": 9.66e-06,
|
1198 |
+
"loss": 3.4032,
|
1199 |
+
"step": 17000
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.171,
|
1203 |
+
"grad_norm": 19.68560791015625,
|
1204 |
+
"learning_rate": 9.658000000000001e-06,
|
1205 |
+
"loss": 3.2144,
|
1206 |
+
"step": 17100
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 0.172,
|
1210 |
+
"grad_norm": 20.55443572998047,
|
1211 |
+
"learning_rate": 9.656e-06,
|
1212 |
+
"loss": 3.347,
|
1213 |
+
"step": 17200
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.173,
|
1217 |
+
"grad_norm": 23.09670639038086,
|
1218 |
+
"learning_rate": 9.654000000000001e-06,
|
1219 |
+
"loss": 3.2204,
|
1220 |
+
"step": 17300
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.174,
|
1224 |
+
"grad_norm": 21.916152954101562,
|
1225 |
+
"learning_rate": 9.652e-06,
|
1226 |
+
"loss": 3.1898,
|
1227 |
+
"step": 17400
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.175,
|
1231 |
+
"grad_norm": 15.10058879852295,
|
1232 |
+
"learning_rate": 9.65e-06,
|
1233 |
+
"loss": 3.3174,
|
1234 |
+
"step": 17500
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 0.176,
|
1238 |
+
"grad_norm": 18.47793197631836,
|
1239 |
+
"learning_rate": 9.648000000000001e-06,
|
1240 |
+
"loss": 3.152,
|
1241 |
+
"step": 17600
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.177,
|
1245 |
+
"grad_norm": 20.482669830322266,
|
1246 |
+
"learning_rate": 9.646e-06,
|
1247 |
+
"loss": 3.2341,
|
1248 |
+
"step": 17700
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 0.178,
|
1252 |
+
"grad_norm": 17.341407775878906,
|
1253 |
+
"learning_rate": 9.644000000000001e-06,
|
1254 |
+
"loss": 3.2945,
|
1255 |
+
"step": 17800
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.179,
|
1259 |
+
"grad_norm": 25.537378311157227,
|
1260 |
+
"learning_rate": 9.642e-06,
|
1261 |
+
"loss": 3.3279,
|
1262 |
+
"step": 17900
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.18,
|
1266 |
+
"grad_norm": 25.134294509887695,
|
1267 |
+
"learning_rate": 9.640000000000001e-06,
|
1268 |
+
"loss": 3.4005,
|
1269 |
+
"step": 18000
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.181,
|
1273 |
+
"grad_norm": 14.844265937805176,
|
1274 |
+
"learning_rate": 9.638e-06,
|
1275 |
+
"loss": 3.2125,
|
1276 |
+
"step": 18100
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 0.182,
|
1280 |
+
"grad_norm": 12.517401695251465,
|
1281 |
+
"learning_rate": 9.636e-06,
|
1282 |
+
"loss": 3.4222,
|
1283 |
+
"step": 18200
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.183,
|
1287 |
+
"grad_norm": 13.91508674621582,
|
1288 |
+
"learning_rate": 9.634000000000001e-06,
|
1289 |
+
"loss": 3.2464,
|
1290 |
+
"step": 18300
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 0.184,
|
1294 |
+
"grad_norm": 20.34067726135254,
|
1295 |
+
"learning_rate": 9.632e-06,
|
1296 |
+
"loss": 3.1909,
|
1297 |
+
"step": 18400
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.185,
|
1301 |
+
"grad_norm": 20.126605987548828,
|
1302 |
+
"learning_rate": 9.630000000000001e-06,
|
1303 |
+
"loss": 3.1106,
|
1304 |
+
"step": 18500
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.186,
|
1308 |
+
"grad_norm": 20.69412612915039,
|
1309 |
+
"learning_rate": 9.628e-06,
|
1310 |
+
"loss": 3.2615,
|
1311 |
+
"step": 18600
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.187,
|
1315 |
+
"grad_norm": 29.957561492919922,
|
1316 |
+
"learning_rate": 9.626e-06,
|
1317 |
+
"loss": 3.2019,
|
1318 |
+
"step": 18700
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 0.188,
|
1322 |
+
"grad_norm": 12.36296558380127,
|
1323 |
+
"learning_rate": 9.624000000000001e-06,
|
1324 |
+
"loss": 3.2337,
|
1325 |
+
"step": 18800
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.189,
|
1329 |
+
"grad_norm": 13.685921669006348,
|
1330 |
+
"learning_rate": 9.622000000000002e-06,
|
1331 |
+
"loss": 3.282,
|
1332 |
+
"step": 18900
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 0.19,
|
1336 |
+
"grad_norm": 20.060331344604492,
|
1337 |
+
"learning_rate": 9.620000000000001e-06,
|
1338 |
+
"loss": 3.1947,
|
1339 |
+
"step": 19000
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.191,
|
1343 |
+
"grad_norm": 13.936528205871582,
|
1344 |
+
"learning_rate": 9.618e-06,
|
1345 |
+
"loss": 3.174,
|
1346 |
+
"step": 19100
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 0.192,
|
1350 |
+
"grad_norm": 21.87002944946289,
|
1351 |
+
"learning_rate": 9.616e-06,
|
1352 |
+
"loss": 3.2355,
|
1353 |
+
"step": 19200
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.193,
|
1357 |
+
"grad_norm": 24.834264755249023,
|
1358 |
+
"learning_rate": 9.614000000000001e-06,
|
1359 |
+
"loss": 3.1823,
|
1360 |
+
"step": 19300
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 0.194,
|
1364 |
+
"grad_norm": 30.83481216430664,
|
1365 |
+
"learning_rate": 9.612000000000002e-06,
|
1366 |
+
"loss": 3.1645,
|
1367 |
+
"step": 19400
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.195,
|
1371 |
+
"grad_norm": 18.5225830078125,
|
1372 |
+
"learning_rate": 9.610000000000001e-06,
|
1373 |
+
"loss": 3.2636,
|
1374 |
+
"step": 19500
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 0.196,
|
1378 |
+
"grad_norm": 27.33846664428711,
|
1379 |
+
"learning_rate": 9.608e-06,
|
1380 |
+
"loss": 3.1772,
|
1381 |
+
"step": 19600
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.197,
|
1385 |
+
"grad_norm": 21.9919490814209,
|
1386 |
+
"learning_rate": 9.606000000000001e-06,
|
1387 |
+
"loss": 3.229,
|
1388 |
+
"step": 19700
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.198,
|
1392 |
+
"grad_norm": 19.65387725830078,
|
1393 |
+
"learning_rate": 9.604000000000002e-06,
|
1394 |
+
"loss": 3.1524,
|
1395 |
+
"step": 19800
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.199,
|
1399 |
+
"grad_norm": 18.683229446411133,
|
1400 |
+
"learning_rate": 9.602e-06,
|
1401 |
+
"loss": 3.4325,
|
1402 |
+
"step": 19900
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 0.2,
|
1406 |
+
"grad_norm": 16.26070785522461,
|
1407 |
+
"learning_rate": 9.600000000000001e-06,
|
1408 |
+
"loss": 3.1583,
|
1409 |
+
"step": 20000
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 0.201,
|
1413 |
+
"grad_norm": 17.30815887451172,
|
1414 |
+
"learning_rate": 9.598e-06,
|
1415 |
+
"loss": 3.1904,
|
1416 |
+
"step": 20100
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 0.202,
|
1420 |
+
"grad_norm": 28.912694931030273,
|
1421 |
+
"learning_rate": 9.596000000000001e-06,
|
1422 |
+
"loss": 3.0794,
|
1423 |
+
"step": 20200
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 0.203,
|
1427 |
+
"grad_norm": 20.792774200439453,
|
1428 |
+
"learning_rate": 9.594000000000002e-06,
|
1429 |
+
"loss": 3.2403,
|
1430 |
+
"step": 20300
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.204,
|
1434 |
+
"grad_norm": 22.178218841552734,
|
1435 |
+
"learning_rate": 9.592e-06,
|
1436 |
+
"loss": 3.1408,
|
1437 |
+
"step": 20400
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 0.205,
|
1441 |
+
"grad_norm": 15.090167045593262,
|
1442 |
+
"learning_rate": 9.59e-06,
|
1443 |
+
"loss": 3.1513,
|
1444 |
+
"step": 20500
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 0.206,
|
1448 |
+
"grad_norm": 19.66379737854004,
|
1449 |
+
"learning_rate": 9.588e-06,
|
1450 |
+
"loss": 3.1574,
|
1451 |
+
"step": 20600
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 0.207,
|
1455 |
+
"grad_norm": 20.961610794067383,
|
1456 |
+
"learning_rate": 9.586000000000001e-06,
|
1457 |
+
"loss": 3.284,
|
1458 |
+
"step": 20700
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 0.208,
|
1462 |
+
"grad_norm": 19.434553146362305,
|
1463 |
+
"learning_rate": 9.584000000000002e-06,
|
1464 |
+
"loss": 3.0802,
|
1465 |
+
"step": 20800
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 0.209,
|
1469 |
+
"grad_norm": 30.214740753173828,
|
1470 |
+
"learning_rate": 9.582e-06,
|
1471 |
+
"loss": 3.2555,
|
1472 |
+
"step": 20900
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 0.21,
|
1476 |
+
"grad_norm": 18.16490364074707,
|
1477 |
+
"learning_rate": 9.58e-06,
|
1478 |
+
"loss": 3.2179,
|
1479 |
+
"step": 21000
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 0.211,
|
1483 |
+
"grad_norm": 22.568527221679688,
|
1484 |
+
"learning_rate": 9.578e-06,
|
1485 |
+
"loss": 3.2268,
|
1486 |
+
"step": 21100
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 0.212,
|
1490 |
+
"grad_norm": 20.349346160888672,
|
1491 |
+
"learning_rate": 9.576000000000001e-06,
|
1492 |
+
"loss": 3.159,
|
1493 |
+
"step": 21200
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 0.213,
|
1497 |
+
"grad_norm": 23.45667266845703,
|
1498 |
+
"learning_rate": 9.574000000000002e-06,
|
1499 |
+
"loss": 3.1728,
|
1500 |
+
"step": 21300
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 0.214,
|
1504 |
+
"grad_norm": 20.883718490600586,
|
1505 |
+
"learning_rate": 9.572000000000001e-06,
|
1506 |
+
"loss": 3.1258,
|
1507 |
+
"step": 21400
|
1508 |
+
},
|
1509 |
+
{
|
1510 |
+
"epoch": 0.215,
|
1511 |
+
"grad_norm": 25.16787338256836,
|
1512 |
+
"learning_rate": 9.57e-06,
|
1513 |
+
"loss": 3.0726,
|
1514 |
+
"step": 21500
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 0.216,
|
1518 |
+
"grad_norm": 21.36046028137207,
|
1519 |
+
"learning_rate": 9.568e-06,
|
1520 |
+
"loss": 3.1267,
|
1521 |
+
"step": 21600
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 0.217,
|
1525 |
+
"grad_norm": 26.431421279907227,
|
1526 |
+
"learning_rate": 9.566000000000001e-06,
|
1527 |
+
"loss": 3.1588,
|
1528 |
+
"step": 21700
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 0.218,
|
1532 |
+
"grad_norm": 27.33740997314453,
|
1533 |
+
"learning_rate": 9.564e-06,
|
1534 |
+
"loss": 3.0679,
|
1535 |
+
"step": 21800
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 0.219,
|
1539 |
+
"grad_norm": 17.818220138549805,
|
1540 |
+
"learning_rate": 9.562000000000001e-06,
|
1541 |
+
"loss": 3.1104,
|
1542 |
+
"step": 21900
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 0.22,
|
1546 |
+
"grad_norm": 25.339937210083008,
|
1547 |
+
"learning_rate": 9.56e-06,
|
1548 |
+
"loss": 3.2342,
|
1549 |
+
"step": 22000
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 0.221,
|
1553 |
+
"grad_norm": 19.325305938720703,
|
1554 |
+
"learning_rate": 9.558e-06,
|
1555 |
+
"loss": 3.1085,
|
1556 |
+
"step": 22100
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 0.222,
|
1560 |
+
"grad_norm": 19.849441528320312,
|
1561 |
+
"learning_rate": 9.556000000000001e-06,
|
1562 |
+
"loss": 3.2078,
|
1563 |
+
"step": 22200
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 0.223,
|
1567 |
+
"grad_norm": 22.334917068481445,
|
1568 |
+
"learning_rate": 9.554e-06,
|
1569 |
+
"loss": 3.1536,
|
1570 |
+
"step": 22300
|
1571 |
+
},
|
1572 |
+
{
|
1573 |
+
"epoch": 0.224,
|
1574 |
+
"grad_norm": 16.38900375366211,
|
1575 |
+
"learning_rate": 9.552000000000001e-06,
|
1576 |
+
"loss": 3.106,
|
1577 |
+
"step": 22400
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 0.225,
|
1581 |
+
"grad_norm": 24.00871467590332,
|
1582 |
+
"learning_rate": 9.55e-06,
|
1583 |
+
"loss": 3.1776,
|
1584 |
+
"step": 22500
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 0.226,
|
1588 |
+
"grad_norm": 19.4804744720459,
|
1589 |
+
"learning_rate": 9.548e-06,
|
1590 |
+
"loss": 3.164,
|
1591 |
+
"step": 22600
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 0.227,
|
1595 |
+
"grad_norm": 17.325008392333984,
|
1596 |
+
"learning_rate": 9.546000000000001e-06,
|
1597 |
+
"loss": 3.2499,
|
1598 |
+
"step": 22700
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 0.228,
|
1602 |
+
"grad_norm": 19.2254695892334,
|
1603 |
+
"learning_rate": 9.544e-06,
|
1604 |
+
"loss": 3.1912,
|
1605 |
+
"step": 22800
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 0.229,
|
1609 |
+
"grad_norm": 19.877927780151367,
|
1610 |
+
"learning_rate": 9.542000000000001e-06,
|
1611 |
+
"loss": 3.1645,
|
1612 |
+
"step": 22900
|
1613 |
+
},
|
1614 |
+
{
|
1615 |
+
"epoch": 0.23,
|
1616 |
+
"grad_norm": 21.79277992248535,
|
1617 |
+
"learning_rate": 9.54e-06,
|
1618 |
+
"loss": 3.0794,
|
1619 |
+
"step": 23000
|
1620 |
+
}
|
1621 |
+
],
|
1622 |
+
"logging_steps": 100,
|
1623 |
+
"max_steps": 500000,
|
1624 |
+
"num_input_tokens_seen": 0,
|
1625 |
+
"num_train_epochs": 5,
|
1626 |
+
"save_steps": 1000,
|
1627 |
+
"stateful_callbacks": {
|
1628 |
+
"TrainerControl": {
|
1629 |
+
"args": {
|
1630 |
+
"should_epoch_stop": false,
|
1631 |
+
"should_evaluate": false,
|
1632 |
+
"should_log": false,
|
1633 |
+
"should_save": true,
|
1634 |
+
"should_training_stop": false
|
1635 |
+
},
|
1636 |
+
"attributes": {}
|
1637 |
+
}
|
1638 |
+
},
|
1639 |
+
"total_flos": 2.4239469785088e+16,
|
1640 |
+
"train_batch_size": 2,
|
1641 |
+
"trial_name": null,
|
1642 |
+
"trial_params": null
|
1643 |
+
}
|
checkpoint-23000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ade0bd39deba891b5e9da350a7c2d57e7871a737e2a0d6bdcb83eec944e6438d
|
3 |
+
size 4859
|