Commit
•
79f6bbc
1
Parent(s):
813a325
doodle-zero
Browse files- .DS_Store +0 -0
- README.md +11 -33
- all_results.json +9 -9
- config.json +100 -99
- model.safetensors +2 -2
- preprocessor_config.json +0 -15
- pytorch_model.bin +3 -0
- test_results.json +4 -5
- train_results.json +5 -5
- trainer_state.json +352 -352
- training_args.bin +2 -2
.DS_Store
ADDED
|
Binary file (6.15 kB). View file
|
|
|
README.md
CHANGED
|
@@ -1,8 +1,7 @@
|
|
| 1 |
---
|
|
|
|
| 2 |
tags:
|
| 3 |
- generated_from_trainer
|
| 4 |
-
metrics:
|
| 5 |
-
- accuracy
|
| 6 |
model-index:
|
| 7 |
- name: results
|
| 8 |
results: []
|
|
@@ -13,10 +12,14 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# results
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
-
|
| 19 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
@@ -39,38 +42,13 @@ The following hyperparameters were used during training:
|
|
| 39 |
- train_batch_size: 256
|
| 40 |
- eval_batch_size: 256
|
| 41 |
- seed: 42
|
| 42 |
-
- distributed_type: multi-GPU
|
| 43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: linear
|
| 45 |
- num_epochs: 5
|
| 46 |
-
- mixed_precision_training: Native AMP
|
| 47 |
-
|
| 48 |
-
### Training results
|
| 49 |
-
|
| 50 |
-
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 51 |
-
|:-------------:|:------:|:-----:|:---------------:|:--------:|
|
| 52 |
-
| 1.7698 | 0.2844 | 5000 | 1.7124 | 0.5802 |
|
| 53 |
-
| 1.5445 | 0.5689 | 10000 | 1.5021 | 0.6270 |
|
| 54 |
-
| 1.439 | 0.8533 | 15000 | 1.3989 | 0.6520 |
|
| 55 |
-
| 1.3625 | 1.1377 | 20000 | 1.3447 | 0.6647 |
|
| 56 |
-
| 1.3192 | 1.4222 | 25000 | 1.2965 | 0.6756 |
|
| 57 |
-
| 1.3 | 1.7066 | 30000 | 1.2788 | 0.6795 |
|
| 58 |
-
| 1.2695 | 1.9910 | 35000 | 1.2347 | 0.6900 |
|
| 59 |
-
| 1.2297 | 2.2754 | 40000 | 1.2160 | 0.6955 |
|
| 60 |
-
| 1.2144 | 2.5599 | 45000 | 1.1894 | 0.7021 |
|
| 61 |
-
| 1.1945 | 2.8443 | 50000 | 1.1734 | 0.7058 |
|
| 62 |
-
| 1.1551 | 3.1287 | 55000 | 1.1611 | 0.7084 |
|
| 63 |
-
| 1.1471 | 3.4132 | 60000 | 1.1523 | 0.7104 |
|
| 64 |
-
| 1.1301 | 3.6976 | 65000 | 1.1314 | 0.7156 |
|
| 65 |
-
| 1.1286 | 3.9820 | 70000 | 1.1220 | 0.7186 |
|
| 66 |
-
| 1.0898 | 4.2665 | 75000 | 1.1140 | 0.7203 |
|
| 67 |
-
| 1.093 | 4.5509 | 80000 | 1.1040 | 0.7232 |
|
| 68 |
-
| 1.0893 | 4.8353 | 85000 | 1.0986 | 0.7246 |
|
| 69 |
-
|
| 70 |
|
| 71 |
### Framework versions
|
| 72 |
|
| 73 |
-
- Transformers 4.
|
| 74 |
-
- Pytorch 2.2.2
|
| 75 |
- Datasets 2.19.0
|
| 76 |
-
- Tokenizers 0.
|
|
|
|
| 1 |
---
|
| 2 |
+
base_model: laszlokiss27/results
|
| 3 |
tags:
|
| 4 |
- generated_from_trainer
|
|
|
|
|
|
|
| 5 |
model-index:
|
| 6 |
- name: results
|
| 7 |
results: []
|
|
|
|
| 12 |
|
| 13 |
# results
|
| 14 |
|
| 15 |
+
This model is a fine-tuned version of [laszlokiss27/results](https://huggingface.co/laszlokiss27/results) on an unknown dataset.
|
| 16 |
It achieves the following results on the evaluation set:
|
| 17 |
+
- eval_loss: 1.1000
|
| 18 |
+
- eval_accuracy: 0.7236
|
| 19 |
+
- eval_runtime: 831.0467
|
| 20 |
+
- eval_samples_per_second: 300.825
|
| 21 |
+
- eval_steps_per_second: 1.176
|
| 22 |
+
- step: 0
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 42 |
- train_batch_size: 256
|
| 43 |
- eval_batch_size: 256
|
| 44 |
- seed: 42
|
|
|
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
- lr_scheduler_type: linear
|
| 47 |
- num_epochs: 5
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
|
| 49 |
### Framework versions
|
| 50 |
|
| 51 |
+
- Transformers 4.33.2
|
| 52 |
+
- Pytorch 2.2.2
|
| 53 |
- Datasets 2.19.0
|
| 54 |
+
- Tokenizers 0.13.3
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
"eval_accuracy": 0.723616,
|
| 4 |
-
"eval_loss": 1.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second":
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
"eval_accuracy": 0.723616,
|
| 4 |
+
"eval_loss": 1.100019931793213,
|
| 5 |
+
"eval_runtime": 831.0467,
|
| 6 |
+
"eval_samples_per_second": 300.825,
|
| 7 |
+
"eval_steps_per_second": 1.176,
|
| 8 |
+
"total_flos": 1.93274424e+18,
|
| 9 |
+
"train_loss": 0.9357909288237652,
|
| 10 |
+
"train_runtime": 45635.435,
|
| 11 |
+
"train_samples_per_second": 493.038,
|
| 12 |
+
"train_steps_per_second": 1.926
|
| 13 |
}
|
config.json
CHANGED
|
@@ -1,4 +1,5 @@
|
|
| 1 |
{
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"MobileViTForImageClassification"
|
| 4 |
],
|
|
@@ -24,104 +25,7 @@
|
|
| 24 |
"id2label": {
|
| 25 |
"0": "aircraft carrier",
|
| 26 |
"1": "airplane",
|
| 27 |
-
"2": "alarm clock",
|
| 28 |
-
"3": "ambulance",
|
| 29 |
-
"4": "angel",
|
| 30 |
-
"5": "animal migration",
|
| 31 |
-
"6": "ant",
|
| 32 |
-
"7": "anvil",
|
| 33 |
-
"8": "apple",
|
| 34 |
-
"9": "arm",
|
| 35 |
"10": "asparagus",
|
| 36 |
-
"11": "axe",
|
| 37 |
-
"12": "backpack",
|
| 38 |
-
"13": "banana",
|
| 39 |
-
"14": "bandage",
|
| 40 |
-
"15": "barn",
|
| 41 |
-
"16": "baseball bat",
|
| 42 |
-
"17": "baseball",
|
| 43 |
-
"18": "basket",
|
| 44 |
-
"19": "basketball",
|
| 45 |
-
"20": "bat",
|
| 46 |
-
"21": "bathtub",
|
| 47 |
-
"22": "beach",
|
| 48 |
-
"23": "bear",
|
| 49 |
-
"24": "beard",
|
| 50 |
-
"25": "bed",
|
| 51 |
-
"26": "bee",
|
| 52 |
-
"27": "belt",
|
| 53 |
-
"28": "bench",
|
| 54 |
-
"29": "bicycle",
|
| 55 |
-
"30": "binoculars",
|
| 56 |
-
"31": "bird",
|
| 57 |
-
"32": "birthday cake",
|
| 58 |
-
"33": "blackberry",
|
| 59 |
-
"34": "blueberry",
|
| 60 |
-
"35": "book",
|
| 61 |
-
"36": "boomerang",
|
| 62 |
-
"37": "bottlecap",
|
| 63 |
-
"38": "bowtie",
|
| 64 |
-
"39": "bracelet",
|
| 65 |
-
"40": "brain",
|
| 66 |
-
"41": "bread",
|
| 67 |
-
"42": "bridge",
|
| 68 |
-
"43": "broccoli",
|
| 69 |
-
"44": "broom",
|
| 70 |
-
"45": "bucket",
|
| 71 |
-
"46": "bulldozer",
|
| 72 |
-
"47": "bus",
|
| 73 |
-
"48": "bush",
|
| 74 |
-
"49": "butterfly",
|
| 75 |
-
"50": "cactus",
|
| 76 |
-
"51": "cake",
|
| 77 |
-
"52": "calculator",
|
| 78 |
-
"53": "calendar",
|
| 79 |
-
"54": "camel",
|
| 80 |
-
"55": "camera",
|
| 81 |
-
"56": "camouflage",
|
| 82 |
-
"57": "campfire",
|
| 83 |
-
"58": "candle",
|
| 84 |
-
"59": "cannon",
|
| 85 |
-
"60": "canoe",
|
| 86 |
-
"61": "car",
|
| 87 |
-
"62": "carrot",
|
| 88 |
-
"63": "castle",
|
| 89 |
-
"64": "cat",
|
| 90 |
-
"65": "ceiling fan",
|
| 91 |
-
"66": "cell phone",
|
| 92 |
-
"67": "cello",
|
| 93 |
-
"68": "chair",
|
| 94 |
-
"69": "chandelier",
|
| 95 |
-
"70": "church",
|
| 96 |
-
"71": "circle",
|
| 97 |
-
"72": "clarinet",
|
| 98 |
-
"73": "clock",
|
| 99 |
-
"74": "cloud",
|
| 100 |
-
"75": "coffee cup",
|
| 101 |
-
"76": "compass",
|
| 102 |
-
"77": "computer",
|
| 103 |
-
"78": "cookie",
|
| 104 |
-
"79": "cooler",
|
| 105 |
-
"80": "couch",
|
| 106 |
-
"81": "cow",
|
| 107 |
-
"82": "crab",
|
| 108 |
-
"83": "crayon",
|
| 109 |
-
"84": "crocodile",
|
| 110 |
-
"85": "crown",
|
| 111 |
-
"86": "cruise ship",
|
| 112 |
-
"87": "cup",
|
| 113 |
-
"88": "diamond",
|
| 114 |
-
"89": "dishwasher",
|
| 115 |
-
"90": "diving board",
|
| 116 |
-
"91": "dog",
|
| 117 |
-
"92": "dolphin",
|
| 118 |
-
"93": "donut",
|
| 119 |
-
"94": "door",
|
| 120 |
-
"95": "dragon",
|
| 121 |
-
"96": "dresser",
|
| 122 |
-
"97": "drill",
|
| 123 |
-
"98": "drums",
|
| 124 |
-
"99": "duck",
|
| 125 |
"100": "dumbbell",
|
| 126 |
"101": "ear",
|
| 127 |
"102": "elbow",
|
|
@@ -132,6 +36,7 @@
|
|
| 132 |
"107": "eyeglasses",
|
| 133 |
"108": "face",
|
| 134 |
"109": "fan",
|
|
|
|
| 135 |
"110": "feather",
|
| 136 |
"111": "fence",
|
| 137 |
"112": "finger",
|
|
@@ -142,6 +47,7 @@
|
|
| 142 |
"117": "flamingo",
|
| 143 |
"118": "flashlight",
|
| 144 |
"119": "flip flops",
|
|
|
|
| 145 |
"120": "floor lamp",
|
| 146 |
"121": "flower",
|
| 147 |
"122": "flying saucer",
|
|
@@ -152,6 +58,7 @@
|
|
| 152 |
"127": "garden hose",
|
| 153 |
"128": "garden",
|
| 154 |
"129": "giraffe",
|
|
|
|
| 155 |
"130": "goatee",
|
| 156 |
"131": "golf club",
|
| 157 |
"132": "grapes",
|
|
@@ -162,6 +69,7 @@
|
|
| 162 |
"137": "hand",
|
| 163 |
"138": "harp",
|
| 164 |
"139": "hat",
|
|
|
|
| 165 |
"140": "headphones",
|
| 166 |
"141": "hedgehog",
|
| 167 |
"142": "helicopter",
|
|
@@ -172,6 +80,7 @@
|
|
| 172 |
"147": "horse",
|
| 173 |
"148": "hospital",
|
| 174 |
"149": "hot air balloon",
|
|
|
|
| 175 |
"150": "hot dog",
|
| 176 |
"151": "hot tub",
|
| 177 |
"152": "hourglass",
|
|
@@ -182,6 +91,7 @@
|
|
| 182 |
"157": "jacket",
|
| 183 |
"158": "jail",
|
| 184 |
"159": "kangaroo",
|
|
|
|
| 185 |
"160": "key",
|
| 186 |
"161": "keyboard",
|
| 187 |
"162": "knee",
|
|
@@ -192,6 +102,7 @@
|
|
| 192 |
"167": "leaf",
|
| 193 |
"168": "leg",
|
| 194 |
"169": "light bulb",
|
|
|
|
| 195 |
"170": "lighter",
|
| 196 |
"171": "lighthouse",
|
| 197 |
"172": "lightning",
|
|
@@ -202,6 +113,7 @@
|
|
| 202 |
"177": "lollipop",
|
| 203 |
"178": "mailbox",
|
| 204 |
"179": "map",
|
|
|
|
| 205 |
"180": "marker",
|
| 206 |
"181": "matches",
|
| 207 |
"182": "megaphone",
|
|
@@ -212,6 +124,7 @@
|
|
| 212 |
"187": "moon",
|
| 213 |
"188": "mosquito",
|
| 214 |
"189": "motorbike",
|
|
|
|
| 215 |
"190": "mountain",
|
| 216 |
"191": "mouse",
|
| 217 |
"192": "moustache",
|
|
@@ -222,6 +135,8 @@
|
|
| 222 |
"197": "necklace",
|
| 223 |
"198": "nose",
|
| 224 |
"199": "ocean",
|
|
|
|
|
|
|
| 225 |
"200": "octagon",
|
| 226 |
"201": "octopus",
|
| 227 |
"202": "onion",
|
|
@@ -232,6 +147,7 @@
|
|
| 232 |
"207": "palm tree",
|
| 233 |
"208": "panda",
|
| 234 |
"209": "pants",
|
|
|
|
| 235 |
"210": "paper clip",
|
| 236 |
"211": "parachute",
|
| 237 |
"212": "parrot",
|
|
@@ -242,6 +158,7 @@
|
|
| 242 |
"217": "pencil",
|
| 243 |
"218": "penguin",
|
| 244 |
"219": "piano",
|
|
|
|
| 245 |
"220": "pickup truck",
|
| 246 |
"221": "picture frame",
|
| 247 |
"222": "pig",
|
|
@@ -252,6 +169,7 @@
|
|
| 252 |
"227": "police car",
|
| 253 |
"228": "pond",
|
| 254 |
"229": "pool",
|
|
|
|
| 255 |
"230": "popsicle",
|
| 256 |
"231": "postcard",
|
| 257 |
"232": "potato",
|
|
@@ -262,6 +180,7 @@
|
|
| 262 |
"237": "radio",
|
| 263 |
"238": "rain",
|
| 264 |
"239": "rainbow",
|
|
|
|
| 265 |
"240": "rake",
|
| 266 |
"241": "remote control",
|
| 267 |
"242": "rhinoceros",
|
|
@@ -272,6 +191,7 @@
|
|
| 272 |
"247": "sailboat",
|
| 273 |
"248": "sandwich",
|
| 274 |
"249": "saw",
|
|
|
|
| 275 |
"250": "saxophone",
|
| 276 |
"251": "school bus",
|
| 277 |
"252": "scissors",
|
|
@@ -282,6 +202,7 @@
|
|
| 282 |
"257": "shark",
|
| 283 |
"258": "sheep",
|
| 284 |
"259": "shoe",
|
|
|
|
| 285 |
"260": "shorts",
|
| 286 |
"261": "shovel",
|
| 287 |
"262": "sink",
|
|
@@ -292,6 +213,7 @@
|
|
| 292 |
"267": "smiley face",
|
| 293 |
"268": "snail",
|
| 294 |
"269": "snake",
|
|
|
|
| 295 |
"270": "snorkel",
|
| 296 |
"271": "snowflake",
|
| 297 |
"272": "snowman",
|
|
@@ -302,6 +224,7 @@
|
|
| 302 |
"277": "spoon",
|
| 303 |
"278": "spreadsheet",
|
| 304 |
"279": "square",
|
|
|
|
| 305 |
"280": "squiggle",
|
| 306 |
"281": "squirrel",
|
| 307 |
"282": "stairs",
|
|
@@ -312,6 +235,7 @@
|
|
| 312 |
"287": "stitches",
|
| 313 |
"288": "stop sign",
|
| 314 |
"289": "stove",
|
|
|
|
| 315 |
"290": "strawberry",
|
| 316 |
"291": "streetlight",
|
| 317 |
"292": "string bean",
|
|
@@ -322,6 +246,8 @@
|
|
| 322 |
"297": "sweater",
|
| 323 |
"298": "swing set",
|
| 324 |
"299": "sword",
|
|
|
|
|
|
|
| 325 |
"300": "syringe",
|
| 326 |
"301": "t-shirt",
|
| 327 |
"302": "table",
|
|
@@ -332,6 +258,7 @@
|
|
| 332 |
"307": "tennis racquet",
|
| 333 |
"308": "tent",
|
| 334 |
"309": "The Eiffel Tower",
|
|
|
|
| 335 |
"310": "The Great Wall of China",
|
| 336 |
"311": "The Mona Lisa",
|
| 337 |
"312": "tiger",
|
|
@@ -342,6 +269,7 @@
|
|
| 342 |
"317": "toothbrush",
|
| 343 |
"318": "toothpaste",
|
| 344 |
"319": "tornado",
|
|
|
|
| 345 |
"320": "tractor",
|
| 346 |
"321": "traffic light",
|
| 347 |
"322": "train",
|
|
@@ -352,6 +280,7 @@
|
|
| 352 |
"327": "trumpet",
|
| 353 |
"328": "umbrella",
|
| 354 |
"329": "underwear",
|
|
|
|
| 355 |
"330": "van",
|
| 356 |
"331": "vase",
|
| 357 |
"332": "violin",
|
|
@@ -362,11 +291,83 @@
|
|
| 362 |
"337": "wheel",
|
| 363 |
"338": "windmill",
|
| 364 |
"339": "wine bottle",
|
|
|
|
| 365 |
"340": "wine glass",
|
| 366 |
"341": "wristwatch",
|
| 367 |
"342": "yoga",
|
| 368 |
"343": "zebra",
|
| 369 |
-
"344": "zigzag"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 370 |
},
|
| 371 |
"ignore_mismatched_sizes": true,
|
| 372 |
"image_size": 64,
|
|
@@ -740,5 +741,5 @@
|
|
| 740 |
"qkv_bias": true,
|
| 741 |
"semantic_loss_ignore_index": 255,
|
| 742 |
"torch_dtype": "float32",
|
| 743 |
-
"transformers_version": "4.
|
| 744 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "laszlokiss27/results",
|
| 3 |
"architectures": [
|
| 4 |
"MobileViTForImageClassification"
|
| 5 |
],
|
|
|
|
| 25 |
"id2label": {
|
| 26 |
"0": "aircraft carrier",
|
| 27 |
"1": "airplane",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
"10": "asparagus",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 29 |
"100": "dumbbell",
|
| 30 |
"101": "ear",
|
| 31 |
"102": "elbow",
|
|
|
|
| 36 |
"107": "eyeglasses",
|
| 37 |
"108": "face",
|
| 38 |
"109": "fan",
|
| 39 |
+
"11": "axe",
|
| 40 |
"110": "feather",
|
| 41 |
"111": "fence",
|
| 42 |
"112": "finger",
|
|
|
|
| 47 |
"117": "flamingo",
|
| 48 |
"118": "flashlight",
|
| 49 |
"119": "flip flops",
|
| 50 |
+
"12": "backpack",
|
| 51 |
"120": "floor lamp",
|
| 52 |
"121": "flower",
|
| 53 |
"122": "flying saucer",
|
|
|
|
| 58 |
"127": "garden hose",
|
| 59 |
"128": "garden",
|
| 60 |
"129": "giraffe",
|
| 61 |
+
"13": "banana",
|
| 62 |
"130": "goatee",
|
| 63 |
"131": "golf club",
|
| 64 |
"132": "grapes",
|
|
|
|
| 69 |
"137": "hand",
|
| 70 |
"138": "harp",
|
| 71 |
"139": "hat",
|
| 72 |
+
"14": "bandage",
|
| 73 |
"140": "headphones",
|
| 74 |
"141": "hedgehog",
|
| 75 |
"142": "helicopter",
|
|
|
|
| 80 |
"147": "horse",
|
| 81 |
"148": "hospital",
|
| 82 |
"149": "hot air balloon",
|
| 83 |
+
"15": "barn",
|
| 84 |
"150": "hot dog",
|
| 85 |
"151": "hot tub",
|
| 86 |
"152": "hourglass",
|
|
|
|
| 91 |
"157": "jacket",
|
| 92 |
"158": "jail",
|
| 93 |
"159": "kangaroo",
|
| 94 |
+
"16": "baseball bat",
|
| 95 |
"160": "key",
|
| 96 |
"161": "keyboard",
|
| 97 |
"162": "knee",
|
|
|
|
| 102 |
"167": "leaf",
|
| 103 |
"168": "leg",
|
| 104 |
"169": "light bulb",
|
| 105 |
+
"17": "baseball",
|
| 106 |
"170": "lighter",
|
| 107 |
"171": "lighthouse",
|
| 108 |
"172": "lightning",
|
|
|
|
| 113 |
"177": "lollipop",
|
| 114 |
"178": "mailbox",
|
| 115 |
"179": "map",
|
| 116 |
+
"18": "basket",
|
| 117 |
"180": "marker",
|
| 118 |
"181": "matches",
|
| 119 |
"182": "megaphone",
|
|
|
|
| 124 |
"187": "moon",
|
| 125 |
"188": "mosquito",
|
| 126 |
"189": "motorbike",
|
| 127 |
+
"19": "basketball",
|
| 128 |
"190": "mountain",
|
| 129 |
"191": "mouse",
|
| 130 |
"192": "moustache",
|
|
|
|
| 135 |
"197": "necklace",
|
| 136 |
"198": "nose",
|
| 137 |
"199": "ocean",
|
| 138 |
+
"2": "alarm clock",
|
| 139 |
+
"20": "bat",
|
| 140 |
"200": "octagon",
|
| 141 |
"201": "octopus",
|
| 142 |
"202": "onion",
|
|
|
|
| 147 |
"207": "palm tree",
|
| 148 |
"208": "panda",
|
| 149 |
"209": "pants",
|
| 150 |
+
"21": "bathtub",
|
| 151 |
"210": "paper clip",
|
| 152 |
"211": "parachute",
|
| 153 |
"212": "parrot",
|
|
|
|
| 158 |
"217": "pencil",
|
| 159 |
"218": "penguin",
|
| 160 |
"219": "piano",
|
| 161 |
+
"22": "beach",
|
| 162 |
"220": "pickup truck",
|
| 163 |
"221": "picture frame",
|
| 164 |
"222": "pig",
|
|
|
|
| 169 |
"227": "police car",
|
| 170 |
"228": "pond",
|
| 171 |
"229": "pool",
|
| 172 |
+
"23": "bear",
|
| 173 |
"230": "popsicle",
|
| 174 |
"231": "postcard",
|
| 175 |
"232": "potato",
|
|
|
|
| 180 |
"237": "radio",
|
| 181 |
"238": "rain",
|
| 182 |
"239": "rainbow",
|
| 183 |
+
"24": "beard",
|
| 184 |
"240": "rake",
|
| 185 |
"241": "remote control",
|
| 186 |
"242": "rhinoceros",
|
|
|
|
| 191 |
"247": "sailboat",
|
| 192 |
"248": "sandwich",
|
| 193 |
"249": "saw",
|
| 194 |
+
"25": "bed",
|
| 195 |
"250": "saxophone",
|
| 196 |
"251": "school bus",
|
| 197 |
"252": "scissors",
|
|
|
|
| 202 |
"257": "shark",
|
| 203 |
"258": "sheep",
|
| 204 |
"259": "shoe",
|
| 205 |
+
"26": "bee",
|
| 206 |
"260": "shorts",
|
| 207 |
"261": "shovel",
|
| 208 |
"262": "sink",
|
|
|
|
| 213 |
"267": "smiley face",
|
| 214 |
"268": "snail",
|
| 215 |
"269": "snake",
|
| 216 |
+
"27": "belt",
|
| 217 |
"270": "snorkel",
|
| 218 |
"271": "snowflake",
|
| 219 |
"272": "snowman",
|
|
|
|
| 224 |
"277": "spoon",
|
| 225 |
"278": "spreadsheet",
|
| 226 |
"279": "square",
|
| 227 |
+
"28": "bench",
|
| 228 |
"280": "squiggle",
|
| 229 |
"281": "squirrel",
|
| 230 |
"282": "stairs",
|
|
|
|
| 235 |
"287": "stitches",
|
| 236 |
"288": "stop sign",
|
| 237 |
"289": "stove",
|
| 238 |
+
"29": "bicycle",
|
| 239 |
"290": "strawberry",
|
| 240 |
"291": "streetlight",
|
| 241 |
"292": "string bean",
|
|
|
|
| 246 |
"297": "sweater",
|
| 247 |
"298": "swing set",
|
| 248 |
"299": "sword",
|
| 249 |
+
"3": "ambulance",
|
| 250 |
+
"30": "binoculars",
|
| 251 |
"300": "syringe",
|
| 252 |
"301": "t-shirt",
|
| 253 |
"302": "table",
|
|
|
|
| 258 |
"307": "tennis racquet",
|
| 259 |
"308": "tent",
|
| 260 |
"309": "The Eiffel Tower",
|
| 261 |
+
"31": "bird",
|
| 262 |
"310": "The Great Wall of China",
|
| 263 |
"311": "The Mona Lisa",
|
| 264 |
"312": "tiger",
|
|
|
|
| 269 |
"317": "toothbrush",
|
| 270 |
"318": "toothpaste",
|
| 271 |
"319": "tornado",
|
| 272 |
+
"32": "birthday cake",
|
| 273 |
"320": "tractor",
|
| 274 |
"321": "traffic light",
|
| 275 |
"322": "train",
|
|
|
|
| 280 |
"327": "trumpet",
|
| 281 |
"328": "umbrella",
|
| 282 |
"329": "underwear",
|
| 283 |
+
"33": "blackberry",
|
| 284 |
"330": "van",
|
| 285 |
"331": "vase",
|
| 286 |
"332": "violin",
|
|
|
|
| 291 |
"337": "wheel",
|
| 292 |
"338": "windmill",
|
| 293 |
"339": "wine bottle",
|
| 294 |
+
"34": "blueberry",
|
| 295 |
"340": "wine glass",
|
| 296 |
"341": "wristwatch",
|
| 297 |
"342": "yoga",
|
| 298 |
"343": "zebra",
|
| 299 |
+
"344": "zigzag",
|
| 300 |
+
"35": "book",
|
| 301 |
+
"36": "boomerang",
|
| 302 |
+
"37": "bottlecap",
|
| 303 |
+
"38": "bowtie",
|
| 304 |
+
"39": "bracelet",
|
| 305 |
+
"4": "angel",
|
| 306 |
+
"40": "brain",
|
| 307 |
+
"41": "bread",
|
| 308 |
+
"42": "bridge",
|
| 309 |
+
"43": "broccoli",
|
| 310 |
+
"44": "broom",
|
| 311 |
+
"45": "bucket",
|
| 312 |
+
"46": "bulldozer",
|
| 313 |
+
"47": "bus",
|
| 314 |
+
"48": "bush",
|
| 315 |
+
"49": "butterfly",
|
| 316 |
+
"5": "animal migration",
|
| 317 |
+
"50": "cactus",
|
| 318 |
+
"51": "cake",
|
| 319 |
+
"52": "calculator",
|
| 320 |
+
"53": "calendar",
|
| 321 |
+
"54": "camel",
|
| 322 |
+
"55": "camera",
|
| 323 |
+
"56": "camouflage",
|
| 324 |
+
"57": "campfire",
|
| 325 |
+
"58": "candle",
|
| 326 |
+
"59": "cannon",
|
| 327 |
+
"6": "ant",
|
| 328 |
+
"60": "canoe",
|
| 329 |
+
"61": "car",
|
| 330 |
+
"62": "carrot",
|
| 331 |
+
"63": "castle",
|
| 332 |
+
"64": "cat",
|
| 333 |
+
"65": "ceiling fan",
|
| 334 |
+
"66": "cell phone",
|
| 335 |
+
"67": "cello",
|
| 336 |
+
"68": "chair",
|
| 337 |
+
"69": "chandelier",
|
| 338 |
+
"7": "anvil",
|
| 339 |
+
"70": "church",
|
| 340 |
+
"71": "circle",
|
| 341 |
+
"72": "clarinet",
|
| 342 |
+
"73": "clock",
|
| 343 |
+
"74": "cloud",
|
| 344 |
+
"75": "coffee cup",
|
| 345 |
+
"76": "compass",
|
| 346 |
+
"77": "computer",
|
| 347 |
+
"78": "cookie",
|
| 348 |
+
"79": "cooler",
|
| 349 |
+
"8": "apple",
|
| 350 |
+
"80": "couch",
|
| 351 |
+
"81": "cow",
|
| 352 |
+
"82": "crab",
|
| 353 |
+
"83": "crayon",
|
| 354 |
+
"84": "crocodile",
|
| 355 |
+
"85": "crown",
|
| 356 |
+
"86": "cruise ship",
|
| 357 |
+
"87": "cup",
|
| 358 |
+
"88": "diamond",
|
| 359 |
+
"89": "dishwasher",
|
| 360 |
+
"9": "arm",
|
| 361 |
+
"90": "diving board",
|
| 362 |
+
"91": "dog",
|
| 363 |
+
"92": "dolphin",
|
| 364 |
+
"93": "donut",
|
| 365 |
+
"94": "door",
|
| 366 |
+
"95": "dragon",
|
| 367 |
+
"96": "dresser",
|
| 368 |
+
"97": "drill",
|
| 369 |
+
"98": "drums",
|
| 370 |
+
"99": "duck"
|
| 371 |
},
|
| 372 |
"ignore_mismatched_sizes": true,
|
| 373 |
"image_size": 64,
|
|
|
|
| 741 |
"qkv_bias": true,
|
| 742 |
"semantic_loss_ignore_index": 255,
|
| 743 |
"torch_dtype": "float32",
|
| 744 |
+
"transformers_version": "4.33.2"
|
| 745 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28184813e695da9074eb277c8000be311e12d53352ec1ed2b6b268532b81b323
|
| 3 |
+
size 18360744
|
preprocessor_config.json
CHANGED
|
@@ -1,19 +1,4 @@
|
|
| 1 |
{
|
| 2 |
-
"_valid_processor_keys": [
|
| 3 |
-
"images",
|
| 4 |
-
"segmentation_maps",
|
| 5 |
-
"do_resize",
|
| 6 |
-
"size",
|
| 7 |
-
"resample",
|
| 8 |
-
"do_rescale",
|
| 9 |
-
"rescale_factor",
|
| 10 |
-
"do_center_crop",
|
| 11 |
-
"crop_size",
|
| 12 |
-
"do_flip_channel_order",
|
| 13 |
-
"return_tensors",
|
| 14 |
-
"data_format",
|
| 15 |
-
"input_data_format"
|
| 16 |
-
],
|
| 17 |
"crop_size": {
|
| 18 |
"height": 28,
|
| 19 |
"width": 28
|
|
|
|
| 1 |
{
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
"crop_size": {
|
| 3 |
"height": 28,
|
| 4 |
"width": 28
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:579e9e6dde3a9f639be92e42b626a108dbd9c7b2f4ca7bee37623bca653d4abb
|
| 3 |
+
size 20803638
|
test_results.json
CHANGED
|
@@ -1,8 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 5.0,
|
| 3 |
"eval_accuracy": 0.723616,
|
| 4 |
-
"eval_loss": 1.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
|
|
|
| 2 |
"eval_accuracy": 0.723616,
|
| 3 |
+
"eval_loss": 1.100019931793213,
|
| 4 |
+
"eval_runtime": 831.0467,
|
| 5 |
+
"eval_samples_per_second": 300.825,
|
| 6 |
+
"eval_steps_per_second": 1.176
|
| 7 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 5.0,
|
| 3 |
+
"total_flos": 1.93274424e+18,
|
| 4 |
+
"train_loss": 0.9357909288237652,
|
| 5 |
+
"train_runtime": 45635.435,
|
| 6 |
+
"train_samples_per_second": 493.038,
|
| 7 |
+
"train_steps_per_second": 1.926
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -10,774 +10,774 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.05688605722737357,
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate": 0.
|
| 15 |
-
"loss":
|
| 16 |
"step": 1000
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.11377211445474714,
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate": 0.
|
| 22 |
-
"loss":
|
| 23 |
"step": 2000
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.17065817168212072,
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate": 0.
|
| 29 |
-
"loss": 1.
|
| 30 |
"step": 3000
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.22754422890949427,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate": 0.
|
| 36 |
-
"loss": 1.
|
| 37 |
"step": 4000
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.2844302861368678,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate": 0.
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 5000
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.2844302861368678,
|
| 48 |
-
"eval_accuracy": 0.
|
| 49 |
-
"eval_loss": 1.
|
| 50 |
-
"eval_runtime":
|
| 51 |
-
"eval_samples_per_second":
|
| 52 |
-
"eval_steps_per_second":
|
| 53 |
"step": 5000
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 0.34131634336424144,
|
| 57 |
-
"grad_norm":
|
| 58 |
-
"learning_rate": 0.
|
| 59 |
-
"loss": 1.
|
| 60 |
"step": 6000
|
| 61 |
},
|
| 62 |
{
|
| 63 |
"epoch": 0.398202400591615,
|
| 64 |
-
"grad_norm":
|
| 65 |
-
"learning_rate": 0.
|
| 66 |
-
"loss": 1.
|
| 67 |
"step": 7000
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"epoch": 0.45508845781898855,
|
| 71 |
-
"grad_norm":
|
| 72 |
-
"learning_rate": 0.
|
| 73 |
-
"loss": 1.
|
| 74 |
"step": 8000
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.5119745150463622,
|
| 78 |
-
"grad_norm": 1.
|
| 79 |
-
"learning_rate": 0.
|
| 80 |
-
"loss": 1.
|
| 81 |
"step": 9000
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.5688605722737357,
|
| 85 |
-
"grad_norm":
|
| 86 |
-
"learning_rate": 0.
|
| 87 |
-
"loss": 1.
|
| 88 |
"step": 10000
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 0.5688605722737357,
|
| 92 |
-
"eval_accuracy": 0.
|
| 93 |
-
"eval_loss": 1.
|
| 94 |
-
"eval_runtime":
|
| 95 |
-
"eval_samples_per_second":
|
| 96 |
-
"eval_steps_per_second":
|
| 97 |
"step": 10000
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"epoch": 0.6257466295011093,
|
| 101 |
-
"grad_norm":
|
| 102 |
-
"learning_rate": 0.
|
| 103 |
-
"loss": 1.
|
| 104 |
"step": 11000
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 0.6826326867284829,
|
| 108 |
-
"grad_norm":
|
| 109 |
-
"learning_rate": 0.
|
| 110 |
-
"loss": 1.
|
| 111 |
"step": 12000
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 0.7395187439558564,
|
| 115 |
-
"grad_norm": 1.
|
| 116 |
-
"learning_rate": 0.
|
| 117 |
-
"loss": 1.
|
| 118 |
"step": 13000
|
| 119 |
},
|
| 120 |
{
|
| 121 |
"epoch": 0.79640480118323,
|
| 122 |
-
"grad_norm":
|
| 123 |
-
"learning_rate": 0.
|
| 124 |
-
"loss": 1.
|
| 125 |
"step": 14000
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 0.8532908584106036,
|
| 129 |
-
"grad_norm": 1.
|
| 130 |
-
"learning_rate": 0.
|
| 131 |
-
"loss": 1.
|
| 132 |
"step": 15000
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 0.8532908584106036,
|
| 136 |
-
"eval_accuracy": 0.
|
| 137 |
-
"eval_loss": 1.
|
| 138 |
-
"eval_runtime":
|
| 139 |
-
"eval_samples_per_second":
|
| 140 |
-
"eval_steps_per_second":
|
| 141 |
"step": 15000
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 0.9101769156379771,
|
| 145 |
-
"grad_norm":
|
| 146 |
-
"learning_rate": 0.
|
| 147 |
-
"loss": 1.
|
| 148 |
"step": 16000
|
| 149 |
},
|
| 150 |
{
|
| 151 |
"epoch": 0.9670629728653507,
|
| 152 |
-
"grad_norm":
|
| 153 |
-
"learning_rate": 0.
|
| 154 |
-
"loss": 1.
|
| 155 |
"step": 17000
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 1.0239490300927243,
|
| 159 |
-
"grad_norm": 1.
|
| 160 |
-
"learning_rate": 0.
|
| 161 |
-
"loss": 1.
|
| 162 |
"step": 18000
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 1.0808350873200978,
|
| 166 |
-
"grad_norm":
|
| 167 |
-
"learning_rate": 0.
|
| 168 |
-
"loss":
|
| 169 |
"step": 19000
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 1.1377211445474713,
|
| 173 |
-
"grad_norm":
|
| 174 |
-
"learning_rate": 0.
|
| 175 |
-
"loss": 1.
|
| 176 |
"step": 20000
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 1.1377211445474713,
|
| 180 |
-
"eval_accuracy": 0.
|
| 181 |
-
"eval_loss": 1.
|
| 182 |
-
"eval_runtime":
|
| 183 |
-
"eval_samples_per_second":
|
| 184 |
-
"eval_steps_per_second":
|
| 185 |
"step": 20000
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 1.194607201774845,
|
| 189 |
-
"grad_norm":
|
| 190 |
-
"learning_rate": 0.
|
| 191 |
-
"loss":
|
| 192 |
"step": 21000
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 1.2514932590022185,
|
| 196 |
-
"grad_norm":
|
| 197 |
-
"learning_rate": 0.
|
| 198 |
-
"loss":
|
| 199 |
"step": 22000
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 1.3083793162295922,
|
| 203 |
-
"grad_norm":
|
| 204 |
-
"learning_rate": 0.
|
| 205 |
-
"loss":
|
| 206 |
"step": 23000
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 1.3652653734569657,
|
| 210 |
-
"grad_norm":
|
| 211 |
-
"learning_rate": 0.
|
| 212 |
-
"loss":
|
| 213 |
"step": 24000
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 1.4221514306843392,
|
| 217 |
-
"grad_norm":
|
| 218 |
-
"learning_rate": 0.
|
| 219 |
-
"loss":
|
| 220 |
"step": 25000
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 1.4221514306843392,
|
| 224 |
-
"eval_accuracy": 0.
|
| 225 |
-
"eval_loss": 1.
|
| 226 |
-
"eval_runtime":
|
| 227 |
-
"eval_samples_per_second":
|
| 228 |
-
"eval_steps_per_second":
|
| 229 |
"step": 25000
|
| 230 |
},
|
| 231 |
{
|
| 232 |
"epoch": 1.4790374879117127,
|
| 233 |
-
"grad_norm":
|
| 234 |
-
"learning_rate": 0.
|
| 235 |
-
"loss":
|
| 236 |
"step": 26000
|
| 237 |
},
|
| 238 |
{
|
| 239 |
"epoch": 1.5359235451390862,
|
| 240 |
-
"grad_norm":
|
| 241 |
-
"learning_rate": 0.
|
| 242 |
-
"loss":
|
| 243 |
"step": 27000
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 1.59280960236646,
|
| 247 |
-
"grad_norm":
|
| 248 |
-
"learning_rate": 0.
|
| 249 |
-
"loss":
|
| 250 |
"step": 28000
|
| 251 |
},
|
| 252 |
{
|
| 253 |
"epoch": 1.6496956595938337,
|
| 254 |
-
"grad_norm":
|
| 255 |
-
"learning_rate": 0.
|
| 256 |
-
"loss":
|
| 257 |
"step": 29000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 1.7065817168212072,
|
| 261 |
-
"grad_norm":
|
| 262 |
-
"learning_rate": 0.
|
| 263 |
-
"loss":
|
| 264 |
"step": 30000
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 1.7065817168212072,
|
| 268 |
-
"eval_accuracy": 0.
|
| 269 |
-
"eval_loss":
|
| 270 |
-
"eval_runtime":
|
| 271 |
-
"eval_samples_per_second":
|
| 272 |
-
"eval_steps_per_second":
|
| 273 |
"step": 30000
|
| 274 |
},
|
| 275 |
{
|
| 276 |
"epoch": 1.7634677740485807,
|
| 277 |
-
"grad_norm":
|
| 278 |
-
"learning_rate": 0.
|
| 279 |
-
"loss":
|
| 280 |
"step": 31000
|
| 281 |
},
|
| 282 |
{
|
| 283 |
"epoch": 1.8203538312759542,
|
| 284 |
-
"grad_norm":
|
| 285 |
-
"learning_rate": 0.
|
| 286 |
-
"loss":
|
| 287 |
"step": 32000
|
| 288 |
},
|
| 289 |
{
|
| 290 |
"epoch": 1.8772398885033277,
|
| 291 |
-
"grad_norm":
|
| 292 |
-
"learning_rate": 0.
|
| 293 |
-
"loss":
|
| 294 |
"step": 33000
|
| 295 |
},
|
| 296 |
{
|
| 297 |
"epoch": 1.9341259457307014,
|
| 298 |
-
"grad_norm":
|
| 299 |
-
"learning_rate": 0.
|
| 300 |
-
"loss":
|
| 301 |
"step": 34000
|
| 302 |
},
|
| 303 |
{
|
| 304 |
"epoch": 1.9910120029580751,
|
| 305 |
-
"grad_norm":
|
| 306 |
-
"learning_rate": 0.
|
| 307 |
-
"loss":
|
| 308 |
"step": 35000
|
| 309 |
},
|
| 310 |
{
|
| 311 |
"epoch": 1.9910120029580751,
|
| 312 |
-
"eval_accuracy": 0.
|
| 313 |
-
"eval_loss":
|
| 314 |
-
"eval_runtime":
|
| 315 |
-
"eval_samples_per_second":
|
| 316 |
-
"eval_steps_per_second":
|
| 317 |
"step": 35000
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 2.0478980601854486,
|
| 321 |
-
"grad_norm":
|
| 322 |
-
"learning_rate": 0.
|
| 323 |
-
"loss":
|
| 324 |
"step": 36000
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 2.104784117412822,
|
| 328 |
-
"grad_norm":
|
| 329 |
-
"learning_rate": 0.
|
| 330 |
-
"loss":
|
| 331 |
"step": 37000
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 2.1616701746401956,
|
| 335 |
-
"grad_norm":
|
| 336 |
-
"learning_rate": 0.
|
| 337 |
-
"loss":
|
| 338 |
"step": 38000
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 2.218556231867569,
|
| 342 |
-
"grad_norm":
|
| 343 |
-
"learning_rate": 0.
|
| 344 |
-
"loss":
|
| 345 |
"step": 39000
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 2.2754422890949426,
|
| 349 |
-
"grad_norm":
|
| 350 |
-
"learning_rate": 0.
|
| 351 |
-
"loss":
|
| 352 |
"step": 40000
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 2.2754422890949426,
|
| 356 |
-
"eval_accuracy": 0.
|
| 357 |
-
"eval_loss":
|
| 358 |
-
"eval_runtime":
|
| 359 |
-
"eval_samples_per_second":
|
| 360 |
-
"eval_steps_per_second":
|
| 361 |
"step": 40000
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 2.3323283463223166,
|
| 365 |
-
"grad_norm": 1.
|
| 366 |
-
"learning_rate": 0.
|
| 367 |
-
"loss":
|
| 368 |
"step": 41000
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 2.38921440354969,
|
| 372 |
-
"grad_norm":
|
| 373 |
-
"learning_rate": 0.
|
| 374 |
-
"loss":
|
| 375 |
"step": 42000
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 2.4461004607770636,
|
| 379 |
-
"grad_norm":
|
| 380 |
-
"learning_rate": 0.
|
| 381 |
-
"loss":
|
| 382 |
"step": 43000
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 2.502986518004437,
|
| 386 |
-
"grad_norm":
|
| 387 |
-
"learning_rate": 0.
|
| 388 |
-
"loss":
|
| 389 |
"step": 44000
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 2.5598725752318106,
|
| 393 |
-
"grad_norm":
|
| 394 |
-
"learning_rate": 0.
|
| 395 |
-
"loss":
|
| 396 |
"step": 45000
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 2.5598725752318106,
|
| 400 |
-
"eval_accuracy": 0.
|
| 401 |
-
"eval_loss":
|
| 402 |
-
"eval_runtime":
|
| 403 |
-
"eval_samples_per_second":
|
| 404 |
-
"eval_steps_per_second":
|
| 405 |
"step": 45000
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 2.6167586324591845,
|
| 409 |
-
"grad_norm":
|
| 410 |
-
"learning_rate": 0.
|
| 411 |
-
"loss":
|
| 412 |
"step": 46000
|
| 413 |
},
|
| 414 |
{
|
| 415 |
"epoch": 2.673644689686558,
|
| 416 |
-
"grad_norm":
|
| 417 |
-
"learning_rate": 0.
|
| 418 |
-
"loss":
|
| 419 |
"step": 47000
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 2.7305307469139315,
|
| 423 |
-
"grad_norm":
|
| 424 |
-
"learning_rate": 0.
|
| 425 |
-
"loss":
|
| 426 |
"step": 48000
|
| 427 |
},
|
| 428 |
{
|
| 429 |
"epoch": 2.787416804141305,
|
| 430 |
-
"grad_norm":
|
| 431 |
-
"learning_rate": 0.
|
| 432 |
-
"loss":
|
| 433 |
"step": 49000
|
| 434 |
},
|
| 435 |
{
|
| 436 |
"epoch": 2.8443028613686785,
|
| 437 |
-
"grad_norm":
|
| 438 |
-
"learning_rate": 0.
|
| 439 |
-
"loss":
|
| 440 |
"step": 50000
|
| 441 |
},
|
| 442 |
{
|
| 443 |
"epoch": 2.8443028613686785,
|
| 444 |
-
"eval_accuracy": 0.
|
| 445 |
-
"eval_loss":
|
| 446 |
-
"eval_runtime":
|
| 447 |
-
"eval_samples_per_second":
|
| 448 |
-
"eval_steps_per_second":
|
| 449 |
"step": 50000
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 2.901188918596052,
|
| 453 |
-
"grad_norm":
|
| 454 |
-
"learning_rate": 0.
|
| 455 |
-
"loss":
|
| 456 |
"step": 51000
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 2.9580749758234255,
|
| 460 |
-
"grad_norm":
|
| 461 |
-
"learning_rate": 0.
|
| 462 |
-
"loss":
|
| 463 |
"step": 52000
|
| 464 |
},
|
| 465 |
{
|
| 466 |
"epoch": 3.0149610330507994,
|
| 467 |
-
"grad_norm":
|
| 468 |
-
"learning_rate": 0.
|
| 469 |
-
"loss":
|
| 470 |
"step": 53000
|
| 471 |
},
|
| 472 |
{
|
| 473 |
"epoch": 3.071847090278173,
|
| 474 |
-
"grad_norm":
|
| 475 |
-
"learning_rate": 0.
|
| 476 |
-
"loss":
|
| 477 |
"step": 54000
|
| 478 |
},
|
| 479 |
{
|
| 480 |
"epoch": 3.1287331475055464,
|
| 481 |
-
"grad_norm":
|
| 482 |
-
"learning_rate": 0.
|
| 483 |
-
"loss":
|
| 484 |
"step": 55000
|
| 485 |
},
|
| 486 |
{
|
| 487 |
"epoch": 3.1287331475055464,
|
| 488 |
-
"eval_accuracy": 0.
|
| 489 |
-
"eval_loss":
|
| 490 |
-
"eval_runtime":
|
| 491 |
-
"eval_samples_per_second":
|
| 492 |
-
"eval_steps_per_second":
|
| 493 |
"step": 55000
|
| 494 |
},
|
| 495 |
{
|
| 496 |
"epoch": 3.18561920473292,
|
| 497 |
-
"grad_norm": 1.
|
| 498 |
-
"learning_rate": 0.
|
| 499 |
-
"loss":
|
| 500 |
"step": 56000
|
| 501 |
},
|
| 502 |
{
|
| 503 |
"epoch": 3.2425052619602934,
|
| 504 |
-
"grad_norm": 1.
|
| 505 |
-
"learning_rate": 0.
|
| 506 |
-
"loss":
|
| 507 |
"step": 57000
|
| 508 |
},
|
| 509 |
{
|
| 510 |
"epoch": 3.299391319187667,
|
| 511 |
-
"grad_norm": 1.
|
| 512 |
-
"learning_rate": 0.
|
| 513 |
-
"loss":
|
| 514 |
"step": 58000
|
| 515 |
},
|
| 516 |
{
|
| 517 |
"epoch": 3.356277376415041,
|
| 518 |
-
"grad_norm":
|
| 519 |
-
"learning_rate": 0.
|
| 520 |
-
"loss":
|
| 521 |
"step": 59000
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"epoch": 3.4131634336424144,
|
| 525 |
-
"grad_norm":
|
| 526 |
-
"learning_rate": 0.
|
| 527 |
-
"loss":
|
| 528 |
"step": 60000
|
| 529 |
},
|
| 530 |
{
|
| 531 |
"epoch": 3.4131634336424144,
|
| 532 |
-
"eval_accuracy": 0.
|
| 533 |
-
"eval_loss":
|
| 534 |
-
"eval_runtime":
|
| 535 |
-
"eval_samples_per_second":
|
| 536 |
-
"eval_steps_per_second":
|
| 537 |
"step": 60000
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 3.470049490869788,
|
| 541 |
-
"grad_norm":
|
| 542 |
-
"learning_rate": 0.
|
| 543 |
-
"loss":
|
| 544 |
"step": 61000
|
| 545 |
},
|
| 546 |
{
|
| 547 |
"epoch": 3.5269355480971614,
|
| 548 |
-
"grad_norm":
|
| 549 |
-
"learning_rate": 0.
|
| 550 |
-
"loss":
|
| 551 |
"step": 62000
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 3.583821605324535,
|
| 555 |
-
"grad_norm":
|
| 556 |
-
"learning_rate": 0.
|
| 557 |
-
"loss":
|
| 558 |
"step": 63000
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 3.6407076625519084,
|
| 562 |
-
"grad_norm":
|
| 563 |
-
"learning_rate": 0.
|
| 564 |
-
"loss":
|
| 565 |
"step": 64000
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 3.697593719779282,
|
| 569 |
-
"grad_norm":
|
| 570 |
-
"learning_rate": 0.
|
| 571 |
-
"loss":
|
| 572 |
"step": 65000
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 3.697593719779282,
|
| 576 |
-
"eval_accuracy": 0.
|
| 577 |
-
"eval_loss":
|
| 578 |
-
"eval_runtime":
|
| 579 |
-
"eval_samples_per_second":
|
| 580 |
-
"eval_steps_per_second":
|
| 581 |
"step": 65000
|
| 582 |
},
|
| 583 |
{
|
| 584 |
"epoch": 3.754479777006656,
|
| 585 |
-
"grad_norm":
|
| 586 |
-
"learning_rate": 0.
|
| 587 |
-
"loss":
|
| 588 |
"step": 66000
|
| 589 |
},
|
| 590 |
{
|
| 591 |
"epoch": 3.8113658342340293,
|
| 592 |
-
"grad_norm": 1.
|
| 593 |
-
"learning_rate": 0.
|
| 594 |
-
"loss":
|
| 595 |
"step": 67000
|
| 596 |
},
|
| 597 |
{
|
| 598 |
"epoch": 3.868251891461403,
|
| 599 |
-
"grad_norm":
|
| 600 |
-
"learning_rate": 0.
|
| 601 |
-
"loss":
|
| 602 |
"step": 68000
|
| 603 |
},
|
| 604 |
{
|
| 605 |
"epoch": 3.9251379486887763,
|
| 606 |
-
"grad_norm":
|
| 607 |
-
"learning_rate": 0.
|
| 608 |
-
"loss":
|
| 609 |
"step": 69000
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 3.98202400591615,
|
| 613 |
-
"grad_norm": 1.
|
| 614 |
-
"learning_rate": 0.
|
| 615 |
-
"loss":
|
| 616 |
"step": 70000
|
| 617 |
},
|
| 618 |
{
|
| 619 |
"epoch": 3.98202400591615,
|
| 620 |
-
"eval_accuracy": 0.
|
| 621 |
-
"eval_loss":
|
| 622 |
-
"eval_runtime":
|
| 623 |
-
"eval_samples_per_second":
|
| 624 |
-
"eval_steps_per_second":
|
| 625 |
"step": 70000
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 4.038910063143524,
|
| 629 |
-
"grad_norm":
|
| 630 |
-
"learning_rate": 0.
|
| 631 |
-
"loss":
|
| 632 |
"step": 71000
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 4.095796120370897,
|
| 636 |
-
"grad_norm":
|
| 637 |
-
"learning_rate": 0.
|
| 638 |
-
"loss":
|
| 639 |
"step": 72000
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 4.152682177598271,
|
| 643 |
-
"grad_norm":
|
| 644 |
-
"learning_rate": 0.
|
| 645 |
-
"loss":
|
| 646 |
"step": 73000
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 4.209568234825644,
|
| 650 |
-
"grad_norm":
|
| 651 |
-
"learning_rate": 0.
|
| 652 |
-
"loss":
|
| 653 |
"step": 74000
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 4.266454292053018,
|
| 657 |
-
"grad_norm":
|
| 658 |
-
"learning_rate": 0.
|
| 659 |
-
"loss":
|
| 660 |
"step": 75000
|
| 661 |
},
|
| 662 |
{
|
| 663 |
"epoch": 4.266454292053018,
|
| 664 |
-
"eval_accuracy": 0.
|
| 665 |
-
"eval_loss":
|
| 666 |
-
"eval_runtime":
|
| 667 |
-
"eval_samples_per_second":
|
| 668 |
-
"eval_steps_per_second":
|
| 669 |
"step": 75000
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 4.323340349280391,
|
| 673 |
-
"grad_norm":
|
| 674 |
-
"learning_rate": 0.
|
| 675 |
-
"loss":
|
| 676 |
"step": 76000
|
| 677 |
},
|
| 678 |
{
|
| 679 |
"epoch": 4.380226406507765,
|
| 680 |
-
"grad_norm":
|
| 681 |
-
"learning_rate": 9.
|
| 682 |
-
"loss":
|
| 683 |
"step": 77000
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 4.437112463735138,
|
| 687 |
-
"grad_norm":
|
| 688 |
-
"learning_rate": 9.
|
| 689 |
-
"loss":
|
| 690 |
"step": 78000
|
| 691 |
},
|
| 692 |
{
|
| 693 |
"epoch": 4.493998520962512,
|
| 694 |
-
"grad_norm": 1.
|
| 695 |
-
"learning_rate": 8.
|
| 696 |
-
"loss":
|
| 697 |
"step": 79000
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 4.550884578189885,
|
| 701 |
-
"grad_norm": 1.
|
| 702 |
-
"learning_rate": 7.
|
| 703 |
-
"loss":
|
| 704 |
"step": 80000
|
| 705 |
},
|
| 706 |
{
|
| 707 |
"epoch": 4.550884578189885,
|
| 708 |
-
"eval_accuracy": 0.
|
| 709 |
-
"eval_loss":
|
| 710 |
-
"eval_runtime":
|
| 711 |
-
"eval_samples_per_second":
|
| 712 |
-
"eval_steps_per_second":
|
| 713 |
"step": 80000
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 4.607770635417259,
|
| 717 |
-
"grad_norm": 1.
|
| 718 |
-
"learning_rate": 6.
|
| 719 |
-
"loss":
|
| 720 |
"step": 81000
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 4.664656692644633,
|
| 724 |
-
"grad_norm":
|
| 725 |
-
"learning_rate": 5.
|
| 726 |
-
"loss":
|
| 727 |
"step": 82000
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 4.721542749872007,
|
| 731 |
-
"grad_norm": 1.
|
| 732 |
-
"learning_rate": 4.
|
| 733 |
-
"loss":
|
| 734 |
"step": 83000
|
| 735 |
},
|
| 736 |
{
|
| 737 |
"epoch": 4.77842880709938,
|
| 738 |
-
"grad_norm":
|
| 739 |
-
"learning_rate": 3.
|
| 740 |
-
"loss":
|
| 741 |
"step": 84000
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 4.835314864326754,
|
| 745 |
-
"grad_norm":
|
| 746 |
-
"learning_rate": 2.
|
| 747 |
-
"loss":
|
| 748 |
"step": 85000
|
| 749 |
},
|
| 750 |
{
|
| 751 |
"epoch": 4.835314864326754,
|
| 752 |
-
"eval_accuracy": 0.
|
| 753 |
-
"eval_loss":
|
| 754 |
-
"eval_runtime":
|
| 755 |
-
"eval_samples_per_second":
|
| 756 |
-
"eval_steps_per_second":
|
| 757 |
"step": 85000
|
| 758 |
},
|
| 759 |
{
|
| 760 |
"epoch": 4.892200921554127,
|
| 761 |
-
"grad_norm":
|
| 762 |
-
"learning_rate": 1.
|
| 763 |
-
"loss":
|
| 764 |
"step": 86000
|
| 765 |
},
|
| 766 |
{
|
| 767 |
"epoch": 4.949086978781501,
|
| 768 |
-
"grad_norm": 1.
|
| 769 |
-
"learning_rate": 8.
|
| 770 |
-
"loss":
|
| 771 |
"step": 87000
|
| 772 |
},
|
| 773 |
{
|
| 774 |
"epoch": 5.0,
|
| 775 |
"step": 87895,
|
| 776 |
-
"total_flos":
|
| 777 |
-
"train_loss":
|
| 778 |
-
"train_runtime":
|
| 779 |
-
"train_samples_per_second":
|
| 780 |
-
"train_steps_per_second":
|
| 781 |
}
|
| 782 |
],
|
| 783 |
"logging_steps": 1000,
|
|
@@ -785,7 +785,7 @@
|
|
| 785 |
"num_input_tokens_seen": 0,
|
| 786 |
"num_train_epochs": 5,
|
| 787 |
"save_steps": 5000,
|
| 788 |
-
"total_flos":
|
| 789 |
"train_batch_size": 256,
|
| 790 |
"trial_name": null,
|
| 791 |
"trial_params": null
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.05688605722737357,
|
| 13 |
+
"grad_norm": 2.4062280654907227,
|
| 14 |
+
"learning_rate": 0.0007908982308436202,
|
| 15 |
+
"loss": 2.1918,
|
| 16 |
"step": 1000
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.11377211445474714,
|
| 20 |
+
"grad_norm": 1.431848406791687,
|
| 21 |
+
"learning_rate": 0.0007817964616872405,
|
| 22 |
+
"loss": 1.4818,
|
| 23 |
"step": 2000
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.17065817168212072,
|
| 27 |
+
"grad_norm": 1.5747077465057373,
|
| 28 |
+
"learning_rate": 0.0007726946925308607,
|
| 29 |
+
"loss": 1.3634,
|
| 30 |
"step": 3000
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.22754422890949427,
|
| 34 |
+
"grad_norm": 1.4864206314086914,
|
| 35 |
+
"learning_rate": 0.0007635929233744809,
|
| 36 |
+
"loss": 1.2967,
|
| 37 |
"step": 4000
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.2844302861368678,
|
| 41 |
+
"grad_norm": 1.2000905275344849,
|
| 42 |
+
"learning_rate": 0.0007544911542181011,
|
| 43 |
+
"loss": 1.2574,
|
| 44 |
"step": 5000
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.2844302861368678,
|
| 48 |
+
"eval_accuracy": 0.689128,
|
| 49 |
+
"eval_loss": 1.238457202911377,
|
| 50 |
+
"eval_runtime": 203.0197,
|
| 51 |
+
"eval_samples_per_second": 1231.407,
|
| 52 |
+
"eval_steps_per_second": 4.812,
|
| 53 |
"step": 5000
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 0.34131634336424144,
|
| 57 |
+
"grad_norm": 1.2910780906677246,
|
| 58 |
+
"learning_rate": 0.0007453893850617214,
|
| 59 |
+
"loss": 1.2181,
|
| 60 |
"step": 6000
|
| 61 |
},
|
| 62 |
{
|
| 63 |
"epoch": 0.398202400591615,
|
| 64 |
+
"grad_norm": 1.1383774280548096,
|
| 65 |
+
"learning_rate": 0.0007362876159053416,
|
| 66 |
+
"loss": 1.1863,
|
| 67 |
"step": 7000
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"epoch": 0.45508845781898855,
|
| 71 |
+
"grad_norm": 1.135689616203308,
|
| 72 |
+
"learning_rate": 0.0007271858467489618,
|
| 73 |
+
"loss": 1.1653,
|
| 74 |
"step": 8000
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.5119745150463622,
|
| 78 |
+
"grad_norm": 1.1965036392211914,
|
| 79 |
+
"learning_rate": 0.0007180840775925821,
|
| 80 |
+
"loss": 1.147,
|
| 81 |
"step": 9000
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.5688605722737357,
|
| 85 |
+
"grad_norm": 1.0561026334762573,
|
| 86 |
+
"learning_rate": 0.0007089823084362024,
|
| 87 |
+
"loss": 1.1281,
|
| 88 |
"step": 10000
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 0.5688605722737357,
|
| 92 |
+
"eval_accuracy": 0.715764,
|
| 93 |
+
"eval_loss": 1.1192152500152588,
|
| 94 |
+
"eval_runtime": 128.8781,
|
| 95 |
+
"eval_samples_per_second": 1939.817,
|
| 96 |
+
"eval_steps_per_second": 7.581,
|
| 97 |
"step": 10000
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"epoch": 0.6257466295011093,
|
| 101 |
+
"grad_norm": 0.9711835980415344,
|
| 102 |
+
"learning_rate": 0.0006998805392798226,
|
| 103 |
+
"loss": 1.1232,
|
| 104 |
"step": 11000
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 0.6826326867284829,
|
| 108 |
+
"grad_norm": 0.8913602828979492,
|
| 109 |
+
"learning_rate": 0.0006907787701234428,
|
| 110 |
+
"loss": 1.0988,
|
| 111 |
"step": 12000
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 0.7395187439558564,
|
| 115 |
+
"grad_norm": 1.092698097229004,
|
| 116 |
+
"learning_rate": 0.000681677000967063,
|
| 117 |
+
"loss": 1.0897,
|
| 118 |
"step": 13000
|
| 119 |
},
|
| 120 |
{
|
| 121 |
"epoch": 0.79640480118323,
|
| 122 |
+
"grad_norm": 0.9319038391113281,
|
| 123 |
+
"learning_rate": 0.0006725752318106833,
|
| 124 |
+
"loss": 1.0826,
|
| 125 |
"step": 14000
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 0.8532908584106036,
|
| 129 |
+
"grad_norm": 1.0223675966262817,
|
| 130 |
+
"learning_rate": 0.0006634734626543035,
|
| 131 |
+
"loss": 1.0698,
|
| 132 |
"step": 15000
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 0.8532908584106036,
|
| 136 |
+
"eval_accuracy": 0.728676,
|
| 137 |
+
"eval_loss": 1.0653605461120605,
|
| 138 |
+
"eval_runtime": 128.0826,
|
| 139 |
+
"eval_samples_per_second": 1951.866,
|
| 140 |
+
"eval_steps_per_second": 7.628,
|
| 141 |
"step": 15000
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 0.9101769156379771,
|
| 145 |
+
"grad_norm": 0.8995338678359985,
|
| 146 |
+
"learning_rate": 0.0006543716934979237,
|
| 147 |
+
"loss": 1.0624,
|
| 148 |
"step": 16000
|
| 149 |
},
|
| 150 |
{
|
| 151 |
"epoch": 0.9670629728653507,
|
| 152 |
+
"grad_norm": 0.8418471217155457,
|
| 153 |
+
"learning_rate": 0.0006452699243415439,
|
| 154 |
+
"loss": 1.0538,
|
| 155 |
"step": 17000
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 1.0239490300927243,
|
| 159 |
+
"grad_norm": 1.024624228477478,
|
| 160 |
+
"learning_rate": 0.0006361681551851641,
|
| 161 |
+
"loss": 1.0311,
|
| 162 |
"step": 18000
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 1.0808350873200978,
|
| 166 |
+
"grad_norm": 0.9130891561508179,
|
| 167 |
+
"learning_rate": 0.0006270663860287844,
|
| 168 |
+
"loss": 0.999,
|
| 169 |
"step": 19000
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 1.1377211445474713,
|
| 173 |
+
"grad_norm": 0.8896342515945435,
|
| 174 |
+
"learning_rate": 0.0006179646168724045,
|
| 175 |
+
"loss": 1.0,
|
| 176 |
"step": 20000
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 1.1377211445474713,
|
| 180 |
+
"eval_accuracy": 0.739712,
|
| 181 |
+
"eval_loss": 1.0235533714294434,
|
| 182 |
+
"eval_runtime": 127.2585,
|
| 183 |
+
"eval_samples_per_second": 1964.505,
|
| 184 |
+
"eval_steps_per_second": 7.677,
|
| 185 |
"step": 20000
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 1.194607201774845,
|
| 189 |
+
"grad_norm": 0.7940112948417664,
|
| 190 |
+
"learning_rate": 0.0006088628477160248,
|
| 191 |
+
"loss": 0.9957,
|
| 192 |
"step": 21000
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 1.2514932590022185,
|
| 196 |
+
"grad_norm": 0.9015308618545532,
|
| 197 |
+
"learning_rate": 0.000599761078559645,
|
| 198 |
+
"loss": 0.9967,
|
| 199 |
"step": 22000
|
| 200 |
},
|
| 201 |
{
|
| 202 |
"epoch": 1.3083793162295922,
|
| 203 |
+
"grad_norm": 0.9106078147888184,
|
| 204 |
+
"learning_rate": 0.0005906593094032653,
|
| 205 |
+
"loss": 0.9939,
|
| 206 |
"step": 23000
|
| 207 |
},
|
| 208 |
{
|
| 209 |
"epoch": 1.3652653734569657,
|
| 210 |
+
"grad_norm": 0.9563422203063965,
|
| 211 |
+
"learning_rate": 0.0005815575402468854,
|
| 212 |
+
"loss": 0.9931,
|
| 213 |
"step": 24000
|
| 214 |
},
|
| 215 |
{
|
| 216 |
"epoch": 1.4221514306843392,
|
| 217 |
+
"grad_norm": 0.7646272778511047,
|
| 218 |
+
"learning_rate": 0.0005724557710905057,
|
| 219 |
+
"loss": 0.9774,
|
| 220 |
"step": 25000
|
| 221 |
},
|
| 222 |
{
|
| 223 |
"epoch": 1.4221514306843392,
|
| 224 |
+
"eval_accuracy": 0.743348,
|
| 225 |
+
"eval_loss": 1.0054922103881836,
|
| 226 |
+
"eval_runtime": 127.7729,
|
| 227 |
+
"eval_samples_per_second": 1956.596,
|
| 228 |
+
"eval_steps_per_second": 7.646,
|
| 229 |
"step": 25000
|
| 230 |
},
|
| 231 |
{
|
| 232 |
"epoch": 1.4790374879117127,
|
| 233 |
+
"grad_norm": 0.7779045104980469,
|
| 234 |
+
"learning_rate": 0.000563354001934126,
|
| 235 |
+
"loss": 0.9792,
|
| 236 |
"step": 26000
|
| 237 |
},
|
| 238 |
{
|
| 239 |
"epoch": 1.5359235451390862,
|
| 240 |
+
"grad_norm": 0.8506484627723694,
|
| 241 |
+
"learning_rate": 0.0005542522327777463,
|
| 242 |
+
"loss": 0.9778,
|
| 243 |
"step": 27000
|
| 244 |
},
|
| 245 |
{
|
| 246 |
"epoch": 1.59280960236646,
|
| 247 |
+
"grad_norm": 0.8443676829338074,
|
| 248 |
+
"learning_rate": 0.0005451504636213664,
|
| 249 |
+
"loss": 0.9715,
|
| 250 |
"step": 28000
|
| 251 |
},
|
| 252 |
{
|
| 253 |
"epoch": 1.6496956595938337,
|
| 254 |
+
"grad_norm": 0.9333568215370178,
|
| 255 |
+
"learning_rate": 0.0005360486944649867,
|
| 256 |
+
"loss": 0.9679,
|
| 257 |
"step": 29000
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 1.7065817168212072,
|
| 261 |
+
"grad_norm": 0.9501623511314392,
|
| 262 |
+
"learning_rate": 0.0005269469253086069,
|
| 263 |
+
"loss": 0.9684,
|
| 264 |
"step": 30000
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 1.7065817168212072,
|
| 268 |
+
"eval_accuracy": 0.749276,
|
| 269 |
+
"eval_loss": 0.9812818765640259,
|
| 270 |
+
"eval_runtime": 128.5758,
|
| 271 |
+
"eval_samples_per_second": 1944.379,
|
| 272 |
+
"eval_steps_per_second": 7.599,
|
| 273 |
"step": 30000
|
| 274 |
},
|
| 275 |
{
|
| 276 |
"epoch": 1.7634677740485807,
|
| 277 |
+
"grad_norm": 0.7442188262939453,
|
| 278 |
+
"learning_rate": 0.0005178451561522272,
|
| 279 |
+
"loss": 0.9636,
|
| 280 |
"step": 31000
|
| 281 |
},
|
| 282 |
{
|
| 283 |
"epoch": 1.8203538312759542,
|
| 284 |
+
"grad_norm": 0.7510819435119629,
|
| 285 |
+
"learning_rate": 0.0005087433869958473,
|
| 286 |
+
"loss": 0.9647,
|
| 287 |
"step": 32000
|
| 288 |
},
|
| 289 |
{
|
| 290 |
"epoch": 1.8772398885033277,
|
| 291 |
+
"grad_norm": 0.7448764443397522,
|
| 292 |
+
"learning_rate": 0.0004996416178394676,
|
| 293 |
+
"loss": 0.9591,
|
| 294 |
"step": 33000
|
| 295 |
},
|
| 296 |
{
|
| 297 |
"epoch": 1.9341259457307014,
|
| 298 |
+
"grad_norm": 0.8019358515739441,
|
| 299 |
+
"learning_rate": 0.0004905398486830878,
|
| 300 |
+
"loss": 0.9513,
|
| 301 |
"step": 34000
|
| 302 |
},
|
| 303 |
{
|
| 304 |
"epoch": 1.9910120029580751,
|
| 305 |
+
"grad_norm": 0.9495121240615845,
|
| 306 |
+
"learning_rate": 0.00048143807952670797,
|
| 307 |
+
"loss": 0.9511,
|
| 308 |
"step": 35000
|
| 309 |
},
|
| 310 |
{
|
| 311 |
"epoch": 1.9910120029580751,
|
| 312 |
+
"eval_accuracy": 0.755448,
|
| 313 |
+
"eval_loss": 0.9558805227279663,
|
| 314 |
+
"eval_runtime": 127.8711,
|
| 315 |
+
"eval_samples_per_second": 1955.094,
|
| 316 |
+
"eval_steps_per_second": 7.641,
|
| 317 |
"step": 35000
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 2.0478980601854486,
|
| 321 |
+
"grad_norm": 0.8410281538963318,
|
| 322 |
+
"learning_rate": 0.00047233631037032825,
|
| 323 |
+
"loss": 0.9081,
|
| 324 |
"step": 36000
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 2.104784117412822,
|
| 328 |
+
"grad_norm": 0.8246123194694519,
|
| 329 |
+
"learning_rate": 0.00046323454121394847,
|
| 330 |
+
"loss": 0.8964,
|
| 331 |
"step": 37000
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 2.1616701746401956,
|
| 335 |
+
"grad_norm": 0.9567108154296875,
|
| 336 |
+
"learning_rate": 0.0004541327720575687,
|
| 337 |
+
"loss": 0.8952,
|
| 338 |
"step": 38000
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 2.218556231867569,
|
| 342 |
+
"grad_norm": 0.8104901313781738,
|
| 343 |
+
"learning_rate": 0.0004450310029011889,
|
| 344 |
+
"loss": 0.8925,
|
| 345 |
"step": 39000
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 2.2754422890949426,
|
| 349 |
+
"grad_norm": 0.9034276008605957,
|
| 350 |
+
"learning_rate": 0.0004359292337448092,
|
| 351 |
+
"loss": 0.8998,
|
| 352 |
"step": 40000
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 2.2754422890949426,
|
| 356 |
+
"eval_accuracy": 0.755948,
|
| 357 |
+
"eval_loss": 0.9492226839065552,
|
| 358 |
+
"eval_runtime": 127.8812,
|
| 359 |
+
"eval_samples_per_second": 1954.94,
|
| 360 |
+
"eval_steps_per_second": 7.64,
|
| 361 |
"step": 40000
|
| 362 |
},
|
| 363 |
{
|
| 364 |
"epoch": 2.3323283463223166,
|
| 365 |
+
"grad_norm": 1.3229442834854126,
|
| 366 |
+
"learning_rate": 0.00042682746458842937,
|
| 367 |
+
"loss": 0.8962,
|
| 368 |
"step": 41000
|
| 369 |
},
|
| 370 |
{
|
| 371 |
"epoch": 2.38921440354969,
|
| 372 |
+
"grad_norm": 0.8582925200462341,
|
| 373 |
+
"learning_rate": 0.00041772569543204965,
|
| 374 |
+
"loss": 0.8976,
|
| 375 |
"step": 42000
|
| 376 |
},
|
| 377 |
{
|
| 378 |
"epoch": 2.4461004607770636,
|
| 379 |
+
"grad_norm": 0.8881712555885315,
|
| 380 |
+
"learning_rate": 0.0004086239262756698,
|
| 381 |
+
"loss": 0.8898,
|
| 382 |
"step": 43000
|
| 383 |
},
|
| 384 |
{
|
| 385 |
"epoch": 2.502986518004437,
|
| 386 |
+
"grad_norm": 0.8713961839675903,
|
| 387 |
+
"learning_rate": 0.00039952215711929005,
|
| 388 |
+
"loss": 0.8927,
|
| 389 |
"step": 44000
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 2.5598725752318106,
|
| 393 |
+
"grad_norm": 0.7883007526397705,
|
| 394 |
+
"learning_rate": 0.00039042038796291027,
|
| 395 |
+
"loss": 0.8967,
|
| 396 |
"step": 45000
|
| 397 |
},
|
| 398 |
{
|
| 399 |
"epoch": 2.5598725752318106,
|
| 400 |
+
"eval_accuracy": 0.760028,
|
| 401 |
+
"eval_loss": 0.937300980091095,
|
| 402 |
+
"eval_runtime": 130.0782,
|
| 403 |
+
"eval_samples_per_second": 1921.921,
|
| 404 |
+
"eval_steps_per_second": 7.511,
|
| 405 |
"step": 45000
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 2.6167586324591845,
|
| 409 |
+
"grad_norm": 0.8600155711174011,
|
| 410 |
+
"learning_rate": 0.00038131861880653055,
|
| 411 |
+
"loss": 0.8927,
|
| 412 |
"step": 46000
|
| 413 |
},
|
| 414 |
{
|
| 415 |
"epoch": 2.673644689686558,
|
| 416 |
+
"grad_norm": 0.8501909971237183,
|
| 417 |
+
"learning_rate": 0.0003722168496501508,
|
| 418 |
+
"loss": 0.8913,
|
| 419 |
"step": 47000
|
| 420 |
},
|
| 421 |
{
|
| 422 |
"epoch": 2.7305307469139315,
|
| 423 |
+
"grad_norm": 0.8116582632064819,
|
| 424 |
+
"learning_rate": 0.000363115080493771,
|
| 425 |
+
"loss": 0.8889,
|
| 426 |
"step": 48000
|
| 427 |
},
|
| 428 |
{
|
| 429 |
"epoch": 2.787416804141305,
|
| 430 |
+
"grad_norm": 0.8065186738967896,
|
| 431 |
+
"learning_rate": 0.0003540133113373912,
|
| 432 |
+
"loss": 0.8896,
|
| 433 |
"step": 49000
|
| 434 |
},
|
| 435 |
{
|
| 436 |
"epoch": 2.8443028613686785,
|
| 437 |
+
"grad_norm": 0.9248031973838806,
|
| 438 |
+
"learning_rate": 0.00034491154218101145,
|
| 439 |
+
"loss": 0.8837,
|
| 440 |
"step": 50000
|
| 441 |
},
|
| 442 |
{
|
| 443 |
"epoch": 2.8443028613686785,
|
| 444 |
+
"eval_accuracy": 0.762176,
|
| 445 |
+
"eval_loss": 0.9251159429550171,
|
| 446 |
+
"eval_runtime": 128.4439,
|
| 447 |
+
"eval_samples_per_second": 1946.376,
|
| 448 |
+
"eval_steps_per_second": 7.606,
|
| 449 |
"step": 50000
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 2.901188918596052,
|
| 453 |
+
"grad_norm": 0.8191467523574829,
|
| 454 |
+
"learning_rate": 0.0003358097730246317,
|
| 455 |
+
"loss": 0.878,
|
| 456 |
"step": 51000
|
| 457 |
},
|
| 458 |
{
|
| 459 |
"epoch": 2.9580749758234255,
|
| 460 |
+
"grad_norm": 0.7620063424110413,
|
| 461 |
+
"learning_rate": 0.0003267080038682519,
|
| 462 |
+
"loss": 0.8832,
|
| 463 |
"step": 52000
|
| 464 |
},
|
| 465 |
{
|
| 466 |
"epoch": 3.0149610330507994,
|
| 467 |
+
"grad_norm": 0.8365482687950134,
|
| 468 |
+
"learning_rate": 0.0003176062347118721,
|
| 469 |
+
"loss": 0.8621,
|
| 470 |
"step": 53000
|
| 471 |
},
|
| 472 |
{
|
| 473 |
"epoch": 3.071847090278173,
|
| 474 |
+
"grad_norm": 0.9817807078361511,
|
| 475 |
+
"learning_rate": 0.00030850446555549235,
|
| 476 |
+
"loss": 0.8224,
|
| 477 |
"step": 54000
|
| 478 |
},
|
| 479 |
{
|
| 480 |
"epoch": 3.1287331475055464,
|
| 481 |
+
"grad_norm": 0.847806453704834,
|
| 482 |
+
"learning_rate": 0.00029940269639911263,
|
| 483 |
+
"loss": 0.8253,
|
| 484 |
"step": 55000
|
| 485 |
},
|
| 486 |
{
|
| 487 |
"epoch": 3.1287331475055464,
|
| 488 |
+
"eval_accuracy": 0.76438,
|
| 489 |
+
"eval_loss": 0.9235970973968506,
|
| 490 |
+
"eval_runtime": 126.2531,
|
| 491 |
+
"eval_samples_per_second": 1980.15,
|
| 492 |
+
"eval_steps_per_second": 7.738,
|
| 493 |
"step": 55000
|
| 494 |
},
|
| 495 |
{
|
| 496 |
"epoch": 3.18561920473292,
|
| 497 |
+
"grad_norm": 1.1729530096054077,
|
| 498 |
+
"learning_rate": 0.00029030092724273285,
|
| 499 |
+
"loss": 0.8225,
|
| 500 |
"step": 56000
|
| 501 |
},
|
| 502 |
{
|
| 503 |
"epoch": 3.2425052619602934,
|
| 504 |
+
"grad_norm": 1.0548408031463623,
|
| 505 |
+
"learning_rate": 0.0002811991580863531,
|
| 506 |
+
"loss": 0.821,
|
| 507 |
"step": 57000
|
| 508 |
},
|
| 509 |
{
|
| 510 |
"epoch": 3.299391319187667,
|
| 511 |
+
"grad_norm": 1.0199774503707886,
|
| 512 |
+
"learning_rate": 0.0002720973889299733,
|
| 513 |
+
"loss": 0.8213,
|
| 514 |
"step": 58000
|
| 515 |
},
|
| 516 |
{
|
| 517 |
"epoch": 3.356277376415041,
|
| 518 |
+
"grad_norm": 0.9180177450180054,
|
| 519 |
+
"learning_rate": 0.00026299561977359353,
|
| 520 |
+
"loss": 0.8274,
|
| 521 |
"step": 59000
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"epoch": 3.4131634336424144,
|
| 525 |
+
"grad_norm": 0.9745663404464722,
|
| 526 |
+
"learning_rate": 0.0002538938506172137,
|
| 527 |
+
"loss": 0.8229,
|
| 528 |
"step": 60000
|
| 529 |
},
|
| 530 |
{
|
| 531 |
"epoch": 3.4131634336424144,
|
| 532 |
+
"eval_accuracy": 0.766832,
|
| 533 |
+
"eval_loss": 0.9138370156288147,
|
| 534 |
+
"eval_runtime": 129.2727,
|
| 535 |
+
"eval_samples_per_second": 1933.897,
|
| 536 |
+
"eval_steps_per_second": 7.558,
|
| 537 |
"step": 60000
|
| 538 |
},
|
| 539 |
{
|
| 540 |
"epoch": 3.470049490869788,
|
| 541 |
+
"grad_norm": 0.8708947896957397,
|
| 542 |
+
"learning_rate": 0.0002447920814608339,
|
| 543 |
+
"loss": 0.8256,
|
| 544 |
"step": 61000
|
| 545 |
},
|
| 546 |
{
|
| 547 |
"epoch": 3.5269355480971614,
|
| 548 |
+
"grad_norm": 0.9808185696601868,
|
| 549 |
+
"learning_rate": 0.00023569031230445418,
|
| 550 |
+
"loss": 0.8298,
|
| 551 |
"step": 62000
|
| 552 |
},
|
| 553 |
{
|
| 554 |
"epoch": 3.583821605324535,
|
| 555 |
+
"grad_norm": 0.8228833079338074,
|
| 556 |
+
"learning_rate": 0.0002265885431480744,
|
| 557 |
+
"loss": 0.827,
|
| 558 |
"step": 63000
|
| 559 |
},
|
| 560 |
{
|
| 561 |
"epoch": 3.6407076625519084,
|
| 562 |
+
"grad_norm": 0.9581019878387451,
|
| 563 |
+
"learning_rate": 0.00021748677399169463,
|
| 564 |
+
"loss": 0.8275,
|
| 565 |
"step": 64000
|
| 566 |
},
|
| 567 |
{
|
| 568 |
"epoch": 3.697593719779282,
|
| 569 |
+
"grad_norm": 0.8560314178466797,
|
| 570 |
+
"learning_rate": 0.00020838500483531488,
|
| 571 |
+
"loss": 0.8145,
|
| 572 |
"step": 65000
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 3.697593719779282,
|
| 576 |
+
"eval_accuracy": 0.769172,
|
| 577 |
+
"eval_loss": 0.9042648673057556,
|
| 578 |
+
"eval_runtime": 129.2138,
|
| 579 |
+
"eval_samples_per_second": 1934.778,
|
| 580 |
+
"eval_steps_per_second": 7.561,
|
| 581 |
"step": 65000
|
| 582 |
},
|
| 583 |
{
|
| 584 |
"epoch": 3.754479777006656,
|
| 585 |
+
"grad_norm": 0.8918451070785522,
|
| 586 |
+
"learning_rate": 0.0001992832356789351,
|
| 587 |
+
"loss": 0.819,
|
| 588 |
"step": 66000
|
| 589 |
},
|
| 590 |
{
|
| 591 |
"epoch": 3.8113658342340293,
|
| 592 |
+
"grad_norm": 1.0977294445037842,
|
| 593 |
+
"learning_rate": 0.00019018146652255533,
|
| 594 |
+
"loss": 0.8122,
|
| 595 |
"step": 67000
|
| 596 |
},
|
| 597 |
{
|
| 598 |
"epoch": 3.868251891461403,
|
| 599 |
+
"grad_norm": 0.7856444716453552,
|
| 600 |
+
"learning_rate": 0.00018107969736617555,
|
| 601 |
+
"loss": 0.8225,
|
| 602 |
"step": 68000
|
| 603 |
},
|
| 604 |
{
|
| 605 |
"epoch": 3.9251379486887763,
|
| 606 |
+
"grad_norm": 0.9270259141921997,
|
| 607 |
+
"learning_rate": 0.00017197792820979578,
|
| 608 |
+
"loss": 0.8158,
|
| 609 |
"step": 69000
|
| 610 |
},
|
| 611 |
{
|
| 612 |
"epoch": 3.98202400591615,
|
| 613 |
+
"grad_norm": 1.082774043083191,
|
| 614 |
+
"learning_rate": 0.00016287615905341603,
|
| 615 |
+
"loss": 0.8156,
|
| 616 |
"step": 70000
|
| 617 |
},
|
| 618 |
{
|
| 619 |
"epoch": 3.98202400591615,
|
| 620 |
+
"eval_accuracy": 0.770764,
|
| 621 |
+
"eval_loss": 0.8961142301559448,
|
| 622 |
+
"eval_runtime": 138.0555,
|
| 623 |
+
"eval_samples_per_second": 1810.866,
|
| 624 |
+
"eval_steps_per_second": 7.077,
|
| 625 |
"step": 70000
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 4.038910063143524,
|
| 629 |
+
"grad_norm": 0.909858226776123,
|
| 630 |
+
"learning_rate": 0.00015377438989703626,
|
| 631 |
+
"loss": 0.7785,
|
| 632 |
"step": 71000
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 4.095796120370897,
|
| 636 |
+
"grad_norm": 0.931280791759491,
|
| 637 |
+
"learning_rate": 0.00014467262074065645,
|
| 638 |
+
"loss": 0.7637,
|
| 639 |
"step": 72000
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 4.152682177598271,
|
| 643 |
+
"grad_norm": 0.94422847032547,
|
| 644 |
+
"learning_rate": 0.0001355708515842767,
|
| 645 |
+
"loss": 0.7612,
|
| 646 |
"step": 73000
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 4.209568234825644,
|
| 650 |
+
"grad_norm": 0.9250127077102661,
|
| 651 |
+
"learning_rate": 0.00012646908242789693,
|
| 652 |
+
"loss": 0.7616,
|
| 653 |
"step": 74000
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 4.266454292053018,
|
| 657 |
+
"grad_norm": 0.8467296957969666,
|
| 658 |
+
"learning_rate": 0.00011736731327151716,
|
| 659 |
+
"loss": 0.7557,
|
| 660 |
"step": 75000
|
| 661 |
},
|
| 662 |
{
|
| 663 |
"epoch": 4.266454292053018,
|
| 664 |
+
"eval_accuracy": 0.77204,
|
| 665 |
+
"eval_loss": 0.9022773504257202,
|
| 666 |
+
"eval_runtime": 144.5432,
|
| 667 |
+
"eval_samples_per_second": 1729.587,
|
| 668 |
+
"eval_steps_per_second": 6.759,
|
| 669 |
"step": 75000
|
| 670 |
},
|
| 671 |
{
|
| 672 |
"epoch": 4.323340349280391,
|
| 673 |
+
"grad_norm": 0.8985564708709717,
|
| 674 |
+
"learning_rate": 0.00010826554411513738,
|
| 675 |
+
"loss": 0.7604,
|
| 676 |
"step": 76000
|
| 677 |
},
|
| 678 |
{
|
| 679 |
"epoch": 4.380226406507765,
|
| 680 |
+
"grad_norm": 0.8618564605712891,
|
| 681 |
+
"learning_rate": 9.916377495875762e-05,
|
| 682 |
+
"loss": 0.7632,
|
| 683 |
"step": 77000
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 4.437112463735138,
|
| 687 |
+
"grad_norm": 0.9467126727104187,
|
| 688 |
+
"learning_rate": 9.006200580237784e-05,
|
| 689 |
+
"loss": 0.7614,
|
| 690 |
"step": 78000
|
| 691 |
},
|
| 692 |
{
|
| 693 |
"epoch": 4.493998520962512,
|
| 694 |
+
"grad_norm": 1.0163730382919312,
|
| 695 |
+
"learning_rate": 8.096023664599807e-05,
|
| 696 |
+
"loss": 0.7575,
|
| 697 |
"step": 79000
|
| 698 |
},
|
| 699 |
{
|
| 700 |
"epoch": 4.550884578189885,
|
| 701 |
+
"grad_norm": 1.1194038391113281,
|
| 702 |
+
"learning_rate": 7.18584674896183e-05,
|
| 703 |
+
"loss": 0.7595,
|
| 704 |
"step": 80000
|
| 705 |
},
|
| 706 |
{
|
| 707 |
"epoch": 4.550884578189885,
|
| 708 |
+
"eval_accuracy": 0.772256,
|
| 709 |
+
"eval_loss": 0.897346019744873,
|
| 710 |
+
"eval_runtime": 136.9434,
|
| 711 |
+
"eval_samples_per_second": 1825.571,
|
| 712 |
+
"eval_steps_per_second": 7.134,
|
| 713 |
"step": 80000
|
| 714 |
},
|
| 715 |
{
|
| 716 |
"epoch": 4.607770635417259,
|
| 717 |
+
"grad_norm": 1.0589629411697388,
|
| 718 |
+
"learning_rate": 6.275669833323853e-05,
|
| 719 |
+
"loss": 0.7548,
|
| 720 |
"step": 81000
|
| 721 |
},
|
| 722 |
{
|
| 723 |
"epoch": 4.664656692644633,
|
| 724 |
+
"grad_norm": 0.8540852665901184,
|
| 725 |
+
"learning_rate": 5.365492917685876e-05,
|
| 726 |
+
"loss": 0.7601,
|
| 727 |
"step": 82000
|
| 728 |
},
|
| 729 |
{
|
| 730 |
"epoch": 4.721542749872007,
|
| 731 |
+
"grad_norm": 1.127475380897522,
|
| 732 |
+
"learning_rate": 4.455316002047898e-05,
|
| 733 |
+
"loss": 0.7554,
|
| 734 |
"step": 83000
|
| 735 |
},
|
| 736 |
{
|
| 737 |
"epoch": 4.77842880709938,
|
| 738 |
+
"grad_norm": 0.9464063048362732,
|
| 739 |
+
"learning_rate": 3.545139086409921e-05,
|
| 740 |
+
"loss": 0.756,
|
| 741 |
"step": 84000
|
| 742 |
},
|
| 743 |
{
|
| 744 |
"epoch": 4.835314864326754,
|
| 745 |
+
"grad_norm": 0.9705914855003357,
|
| 746 |
+
"learning_rate": 2.634962170771944e-05,
|
| 747 |
+
"loss": 0.7581,
|
| 748 |
"step": 85000
|
| 749 |
},
|
| 750 |
{
|
| 751 |
"epoch": 4.835314864326754,
|
| 752 |
+
"eval_accuracy": 0.773724,
|
| 753 |
+
"eval_loss": 0.8925997018814087,
|
| 754 |
+
"eval_runtime": 138.7415,
|
| 755 |
+
"eval_samples_per_second": 1801.913,
|
| 756 |
+
"eval_steps_per_second": 7.042,
|
| 757 |
"step": 85000
|
| 758 |
},
|
| 759 |
{
|
| 760 |
"epoch": 4.892200921554127,
|
| 761 |
+
"grad_norm": 0.8879310488700867,
|
| 762 |
+
"learning_rate": 1.7247852551339668e-05,
|
| 763 |
+
"loss": 0.758,
|
| 764 |
"step": 86000
|
| 765 |
},
|
| 766 |
{
|
| 767 |
"epoch": 4.949086978781501,
|
| 768 |
+
"grad_norm": 1.2024400234222412,
|
| 769 |
+
"learning_rate": 8.146083394959896e-06,
|
| 770 |
+
"loss": 0.751,
|
| 771 |
"step": 87000
|
| 772 |
},
|
| 773 |
{
|
| 774 |
"epoch": 5.0,
|
| 775 |
"step": 87895,
|
| 776 |
+
"total_flos": 1.93274424e+18,
|
| 777 |
+
"train_loss": 0.9357909288237652,
|
| 778 |
+
"train_runtime": 45635.435,
|
| 779 |
+
"train_samples_per_second": 493.038,
|
| 780 |
+
"train_steps_per_second": 1.926
|
| 781 |
}
|
| 782 |
],
|
| 783 |
"logging_steps": 1000,
|
|
|
|
| 785 |
"num_input_tokens_seen": 0,
|
| 786 |
"num_train_epochs": 5,
|
| 787 |
"save_steps": 5000,
|
| 788 |
+
"total_flos": 1.93274424e+18,
|
| 789 |
"train_batch_size": 256,
|
| 790 |
"trial_name": null,
|
| 791 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bf823d8cbbdbfa0a6f25c927a43b2dbea4fdd4c372f9f50bf4026741c3ca5e20
|
| 3 |
+
size 4472
|