Hugging Face's logo

luffy19
/
custom_tokenizer

Model card Files
xet
 Community
luffy19 commited on
Commit
c129165
·
verified ·
1 Parent(s): a499a23

Upload hybrid40k/metrics.jsonl with huggingface_hub

Browse files
Files changed (1) hide show
  1. hybrid40k/metrics.jsonl +39 -0
hybrid40k/metrics.jsonl ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"event": "fertility", "docs": 2000, "chars": 7667339, "tokens": 2084475, "tokens_per_1k_chars": 271.86420216974886}
2
+ {"event": "train", "epoch": 0, "step": 20, "loss": 7.015094041824341, "perplexity": 1113.3113391180848, "learning_rate": 0.000299999963404448, "grad_norm": 3.1897075176239014, "tokens_per_sec": 135121.8834151962, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
3
+ {"event": "train", "epoch": 0, "step": 40, "loss": 6.802230000495911, "perplexity": 899.8517256698243, "learning_rate": 0.0002999998536178096, "grad_norm": 1.242748737335205, "tokens_per_sec": 135084.83013719902, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
4
+ {"event": "train", "epoch": 0, "step": 60, "loss": 6.798467397689819, "perplexity": 896.4723027417729, "learning_rate": 0.0002999996706401383, "grad_norm": 2.565810203552246, "tokens_per_sec": 134880.93651106762, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
5
+ {"event": "train", "epoch": 0, "step": 80, "loss": 6.48967170715332, "perplexity": 658.3072101883931, "learning_rate": 0.00029999941447152363, "grad_norm": 0.6995163559913635, "tokens_per_sec": 135083.6352085045, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
6
+ {"event": "train", "epoch": 0, "step": 100, "loss": 6.69958221912384, "perplexity": 812.0664884392409, "learning_rate": 0.00029999908511209046, "grad_norm": 0.9197239875793457, "tokens_per_sec": 134950.33435613845, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
7
+ {"event": "train", "epoch": 0, "step": 120, "loss": 6.934339642524719, "perplexity": 1026.9408799779667, "learning_rate": 0.0002999986825619994, "grad_norm": 1.4646602869033813, "tokens_per_sec": 135073.28004528684, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
8
+ {"event": "train", "epoch": 0, "step": 140, "loss": 6.241546273231506, "perplexity": 513.6521437583104, "learning_rate": 0.0002999982068214471, "grad_norm": 0.9773970246315002, "tokens_per_sec": 135032.0079306364, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
9
+ {"event": "train", "epoch": 0, "step": 160, "loss": 6.230391144752502, "perplexity": 507.9541282356086, "learning_rate": 0.0002999976578906656, "grad_norm": 0.7949209809303284, "tokens_per_sec": 134941.05951923935, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
10
+ {"event": "train", "epoch": 0, "step": 180, "loss": 6.004724383354187, "perplexity": 405.3392550923769, "learning_rate": 0.00029999703576992243, "grad_norm": 1.3869645595550537, "tokens_per_sec": 134932.8457282885, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
11
+ {"event": "train", "epoch": 0, "step": 200, "loss": 6.291043400764465, "perplexity": 539.7161756665872, "learning_rate": 0.00029999634045952154, "grad_norm": 0.9947767853736877, "tokens_per_sec": 135019.27307914858, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
12
+ {"event": "train", "epoch": 0, "step": 220, "loss": 5.905191659927368, "perplexity": 366.93754627190907, "learning_rate": 0.0002999955719598023, "grad_norm": 0.9782528281211853, "tokens_per_sec": 134923.8382040385, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
13
+ {"event": "train", "epoch": 0, "step": 240, "loss": 5.706706166267395, "perplexity": 300.87839092974946, "learning_rate": 0.0002999947302711394, "grad_norm": 1.2111735343933105, "tokens_per_sec": 134978.4316416052, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
14
+ {"event": "train", "epoch": 0, "step": 260, "loss": 5.965428829193115, "perplexity": 389.72011560498925, "learning_rate": 0.00029999381539394374, "grad_norm": 1.0188724994659424, "tokens_per_sec": 134947.81677434559, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
15
+ {"event": "train", "epoch": 0, "step": 280, "loss": 5.780364274978638, "perplexity": 323.87714927284117, "learning_rate": 0.00029999282732866164, "grad_norm": 0.9246698021888733, "tokens_per_sec": 134930.46126659279, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
16
+ {"event": "train", "epoch": 0, "step": 300, "loss": 5.872432351112366, "perplexity": 355.1116869464106, "learning_rate": 0.000299991766075775, "grad_norm": 1.1663424968719482, "tokens_per_sec": 134942.51690918018, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
17
+ {"event": "train", "epoch": 0, "step": 320, "loss": 5.507466197013855, "perplexity": 246.52568750875997, "learning_rate": 0.00029999063163580196, "grad_norm": 1.1033575534820557, "tokens_per_sec": 135014.2328639955, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
18
+ {"event": "train", "epoch": 0, "step": 340, "loss": 5.1011106967926025, "perplexity": 164.20418711632095, "learning_rate": 0.0002999894240092959, "grad_norm": 1.4245985746383667, "tokens_per_sec": 135064.51922404135, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
19
+ {"event": "train", "epoch": 0, "step": 360, "loss": 5.659422516822815, "perplexity": 286.98286691655517, "learning_rate": 0.000299988143196846, "grad_norm": 0.9643398523330688, "tokens_per_sec": 135007.999434188, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
20
+ {"event": "train", "epoch": 0, "step": 380, "loss": 5.160647392272949, "perplexity": 174.27724483339006, "learning_rate": 0.0002999867891990774, "grad_norm": 1.125593662261963, "tokens_per_sec": 134983.7343283025, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
21
+ {"event": "train", "epoch": 0, "step": 400, "loss": 5.57046639919281, "perplexity": 262.55652684017804, "learning_rate": 0.0002999853620166507, "grad_norm": 0.974700391292572, "tokens_per_sec": 135257.38707585356, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
22
+ {"event": "train", "epoch": 0, "step": 420, "loss": 5.667716026306152, "perplexity": 289.3728590570435, "learning_rate": 0.0002999838616502626, "grad_norm": 0.9588457942008972, "tokens_per_sec": 134952.18947651624, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
23
+ {"event": "train", "epoch": 0, "step": 440, "loss": 5.019026160240173, "perplexity": 151.2639252195361, "learning_rate": 0.0002999822881006446, "grad_norm": 1.1147435903549194, "tokens_per_sec": 134982.93889873748, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
24
+ {"event": "train", "epoch": 0, "step": 460, "loss": 5.247012495994568, "perplexity": 189.9978005412062, "learning_rate": 0.0002999806413685647, "grad_norm": 1.1495624780654907, "tokens_per_sec": 135032.405933454, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
25
+ {"event": "train", "epoch": 0, "step": 480, "loss": 5.582234859466553, "perplexity": 265.66466603004125, "learning_rate": 0.00029997892145482646, "grad_norm": 1.1742076873779297, "tokens_per_sec": 135052.839896705, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
26
+ {"event": "train", "epoch": 0, "step": 500, "loss": 4.9827598333358765, "perplexity": 145.876421227539, "learning_rate": 0.0002999771283602691, "grad_norm": 1.2809275388717651, "tokens_per_sec": 135004.94924211391, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
27
+ {"event": "train", "epoch": 0, "step": 520, "loss": 4.99802303314209, "perplexity": 148.1200410435656, "learning_rate": 0.0002999752620857675, "grad_norm": 1.1001681089401245, "tokens_per_sec": 135006.93848218198, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
28
+ {"event": "train", "epoch": 0, "step": 540, "loss": 4.691757559776306, "perplexity": 109.0446640009352, "learning_rate": 0.0002999733226322324, "grad_norm": 1.3323594331741333, "tokens_per_sec": 134983.46918407254, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
29
+ {"event": "train", "epoch": 0, "step": 560, "loss": 4.788676619529724, "perplexity": 120.14226947897541, "learning_rate": 0.0002999713100006102, "grad_norm": 0.9245190620422363, "tokens_per_sec": 135023.91571551235, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
30
+ {"event": "train", "epoch": 0, "step": 580, "loss": 4.975244045257568, "perplexity": 144.7841547233292, "learning_rate": 0.00029996922419188286, "grad_norm": 0.8734283447265625, "tokens_per_sec": 135034.26331095846, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
31
+ {"event": "train", "epoch": 0, "step": 600, "loss": 4.752752065658569, "perplexity": 115.90281817902317, "learning_rate": 0.0002999670652070679, "grad_norm": 0.7748112082481384, "tokens_per_sec": 135011.84552499713, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
32
+ {"event": "train", "epoch": 0, "step": 620, "loss": 5.097843647003174, "perplexity": 163.6685992339066, "learning_rate": 0.00029996483304721894, "grad_norm": 0.961672842502594, "tokens_per_sec": 134974.58745426012, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
33
+ {"event": "train", "epoch": 0, "step": 640, "loss": 4.610710144042969, "perplexity": 100.55553319997271, "learning_rate": 0.0002999625277134251, "grad_norm": 0.9205366969108582, "tokens_per_sec": 134922.77857436665, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
34
+ {"event": "train", "epoch": 0, "step": 660, "loss": 5.085529565811157, "perplexity": 161.66552910992388, "learning_rate": 0.00029996014920681156, "grad_norm": 0.8476035594940186, "tokens_per_sec": 135000.17530511314, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
35
+ {"event": "train", "epoch": 0, "step": 680, "loss": 5.0355952978134155, "perplexity": 153.7911168826435, "learning_rate": 0.0002999576975285388, "grad_norm": 1.2907016277313232, "tokens_per_sec": 135017.68139160436, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
36
+ {"event": "train", "epoch": 0, "step": 700, "loss": 5.023068130016327, "perplexity": 151.8765667386611, "learning_rate": 0.0002999551726798027, "grad_norm": 1.1270508766174316, "tokens_per_sec": 135053.7688601607, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
37
+ {"event": "train", "epoch": 0, "step": 720, "loss": 4.7866517305374146, "perplexity": 119.89924085602638, "learning_rate": 0.00029995257466183546, "grad_norm": 0.7744899392127991, "tokens_per_sec": 135045.67403672123, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
38
+ {"event": "train", "epoch": 0, "step": 740, "loss": 4.666993021965027, "perplexity": 106.3773865575008, "learning_rate": 0.00029994990347590475, "grad_norm": 0.7438130974769592, "tokens_per_sec": 134972.20151707396, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
39
+ {"event": "train", "epoch": 0, "step": 760, "loss": 4.43474018573761, "perplexity": 84.33021184996495, "learning_rate": 0.00029994715912331387, "grad_norm": 0.7224310636520386, "tokens_per_sec": 135022.19127045022, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}