Upload hybrid40k/metrics.jsonl with huggingface_hub
Browse files- hybrid40k/metrics.jsonl +86 -0
hybrid40k/metrics.jsonl
CHANGED
|
@@ -2059,3 +2059,89 @@
|
|
| 2059 |
{"event": "train", "epoch": 1, "step": 41140, "loss": 2.4040462374687195, "perplexity": 11.067869127695182, "learning_rate": 0.00017002878061168038, "grad_norm": 0.9795651435852051, "tokens_per_sec": 134199.51146475234, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2060 |
{"event": "train", "epoch": 1, "step": 41160, "loss": 2.5507077276706696, "perplexity": 12.816170932549133, "learning_rate": 0.0001699249348570827, "grad_norm": 0.7334607243537903, "tokens_per_sec": 134571.10857711313, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2061 |
{"event": "train", "epoch": 1, "step": 41180, "loss": 2.0979593992233276, "perplexity": 8.149523010518797, "learning_rate": 0.0001698210793802984, "grad_norm": 0.6085432171821594, "tokens_per_sec": 134069.9106863624, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2059 |
{"event": "train", "epoch": 1, "step": 41140, "loss": 2.4040462374687195, "perplexity": 11.067869127695182, "learning_rate": 0.00017002878061168038, "grad_norm": 0.9795651435852051, "tokens_per_sec": 134199.51146475234, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2060 |
{"event": "train", "epoch": 1, "step": 41160, "loss": 2.5507077276706696, "perplexity": 12.816170932549133, "learning_rate": 0.0001699249348570827, "grad_norm": 0.7334607243537903, "tokens_per_sec": 134571.10857711313, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2061 |
{"event": "train", "epoch": 1, "step": 41180, "loss": 2.0979593992233276, "perplexity": 8.149523010518797, "learning_rate": 0.0001698210793802984, "grad_norm": 0.6085432171821594, "tokens_per_sec": 134069.9106863624, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2062 |
+
{"event": "train", "epoch": 1, "step": 41200, "loss": 2.08245450258255, "perplexity": 8.024140035451415, "learning_rate": 0.00016971721423200293, "grad_norm": 0.6189667582511902, "tokens_per_sec": 134063.3718326263, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2063 |
+
{"event": "train", "epoch": 1, "step": 41220, "loss": 2.303906559944153, "perplexity": 10.013223404722918, "learning_rate": 0.0001696133394628762, "grad_norm": 0.6335381269454956, "tokens_per_sec": 133235.6356388651, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2064 |
+
{"event": "train", "epoch": 1, "step": 41240, "loss": 2.313007652759552, "perplexity": 10.10477063834264, "learning_rate": 0.0001695094551236031, "grad_norm": 0.6713374853134155, "tokens_per_sec": 134578.7512712325, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2065 |
+
{"event": "train", "epoch": 1, "step": 41260, "loss": 2.551048755645752, "perplexity": 12.820542350716146, "learning_rate": 0.00016940556126487284, "grad_norm": 0.7299486994743347, "tokens_per_sec": 135309.32008321036, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2066 |
+
{"event": "train", "epoch": 1, "step": 41280, "loss": 2.701043516397476, "perplexity": 14.89526707320341, "learning_rate": 0.00016930165793737956, "grad_norm": 1.0057662725448608, "tokens_per_sec": 133928.81320666647, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2067 |
+
{"event": "train", "epoch": 1, "step": 41300, "loss": 2.107017606496811, "perplexity": 8.223678429662726, "learning_rate": 0.00016919774519182182, "grad_norm": 0.5725177526473999, "tokens_per_sec": 134184.57497512823, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2068 |
+
{"event": "train", "epoch": 1, "step": 41320, "loss": 2.3638581931591034, "perplexity": 10.631892316971749, "learning_rate": 0.00016909382307890304, "grad_norm": 0.5999882221221924, "tokens_per_sec": 133831.00344999167, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2069 |
+
{"event": "train", "epoch": 1, "step": 41340, "loss": 2.103885143995285, "perplexity": 8.197958369949616, "learning_rate": 0.0001689898916493309, "grad_norm": 0.7212738990783691, "tokens_per_sec": 134489.59661816366, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2070 |
+
{"event": "train", "epoch": 1, "step": 41360, "loss": 1.9622043371200562, "perplexity": 7.114993636192027, "learning_rate": 0.00016888595095381794, "grad_norm": 0.571167528629303, "tokens_per_sec": 133039.34241822915, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2071 |
+
{"event": "train", "epoch": 1, "step": 41380, "loss": 1.8752244412899017, "perplexity": 6.522282825636505, "learning_rate": 0.00016878200104308085, "grad_norm": 0.5866857767105103, "tokens_per_sec": 133064.3305411298, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2072 |
+
{"event": "train", "epoch": 1, "step": 41400, "loss": 3.0521572828292847, "perplexity": 21.160945361874436, "learning_rate": 0.00016867804196784114, "grad_norm": 0.8663444519042969, "tokens_per_sec": 133749.47422548986, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2073 |
+
{"event": "train", "epoch": 1, "step": 41420, "loss": 2.096329092979431, "perplexity": 8.136247616689083, "learning_rate": 0.0001685740737788248, "grad_norm": 0.5678447484970093, "tokens_per_sec": 133917.19897339758, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2074 |
+
{"event": "train", "epoch": 1, "step": 41440, "loss": 2.383033573627472, "perplexity": 10.837730099049464, "learning_rate": 0.00016847009652676184, "grad_norm": 0.7079859375953674, "tokens_per_sec": 132935.11188145625, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2075 |
+
{"event": "train", "epoch": 1, "step": 41460, "loss": 2.308554530143738, "perplexity": 10.059872897452717, "learning_rate": 0.00016836611026238723, "grad_norm": 0.5851835608482361, "tokens_per_sec": 134165.8435526886, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2076 |
+
{"event": "train", "epoch": 1, "step": 41480, "loss": 2.023071199655533, "perplexity": 7.5615122227729215, "learning_rate": 0.00016826211503644003, "grad_norm": 0.5215581655502319, "tokens_per_sec": 134592.98480823, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2077 |
+
{"event": "train", "epoch": 1, "step": 41500, "loss": 1.9992883801460266, "perplexity": 7.383799770384223, "learning_rate": 0.00016815811089966373, "grad_norm": 0.5530975461006165, "tokens_per_sec": 134608.40787128266, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2078 |
+
{"event": "train", "epoch": 1, "step": 41520, "loss": 1.8826739490032196, "perplexity": 6.5710520496720335, "learning_rate": 0.00016805409790280612, "grad_norm": 0.5422793626785278, "tokens_per_sec": 133835.56472728265, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2079 |
+
{"event": "train", "epoch": 1, "step": 41540, "loss": 2.0201832354068756, "perplexity": 7.53970634825647, "learning_rate": 0.00016795007609661942, "grad_norm": 0.576108455657959, "tokens_per_sec": 131061.12590222026, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2080 |
+
{"event": "train", "epoch": 1, "step": 41560, "loss": 2.1361435651779175, "perplexity": 8.4667232226779, "learning_rate": 0.0001678460455318602, "grad_norm": 0.5538779497146606, "tokens_per_sec": 131417.65916440528, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2081 |
+
{"event": "train", "epoch": 1, "step": 41580, "loss": 1.9794136583805084, "perplexity": 7.238497508281058, "learning_rate": 0.00016774200625928903, "grad_norm": 0.6545062065124512, "tokens_per_sec": 134515.92247658383, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2082 |
+
{"event": "train", "epoch": 1, "step": 41600, "loss": 2.163272738456726, "perplexity": 8.699562512326944, "learning_rate": 0.00016763795832967104, "grad_norm": 0.5859863758087158, "tokens_per_sec": 133798.04117553143, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2083 |
+
{"event": "train", "epoch": 1, "step": 41620, "loss": 1.9813221096992493, "perplexity": 7.252325018764884, "learning_rate": 0.00016753390179377546, "grad_norm": 0.562808632850647, "tokens_per_sec": 133569.96584143693, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2084 |
+
{"event": "train", "epoch": 1, "step": 41640, "loss": 1.9305072128772736, "perplexity": 6.893005576258375, "learning_rate": 0.0001674298367023756, "grad_norm": 0.6139111518859863, "tokens_per_sec": 135269.36807740649, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2085 |
+
{"event": "train", "epoch": 1, "step": 41660, "loss": 1.9897716641426086, "perplexity": 7.313863554327116, "learning_rate": 0.0001673257631062492, "grad_norm": 0.5318440198898315, "tokens_per_sec": 134739.44861891994, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2086 |
+
{"event": "train", "epoch": 1, "step": 41680, "loss": 2.0676227807998657, "perplexity": 7.906006450708622, "learning_rate": 0.00016722168105617787, "grad_norm": 0.5623111128807068, "tokens_per_sec": 134114.52268371606, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2087 |
+
{"event": "train", "epoch": 1, "step": 41700, "loss": 2.796603500843048, "perplexity": 16.388887289323463, "learning_rate": 0.0001671175906029475, "grad_norm": 0.7473816275596619, "tokens_per_sec": 134955.1047687315, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2088 |
+
{"event": "train", "epoch": 1, "step": 41720, "loss": 1.8933384418487549, "perplexity": 6.641503986616922, "learning_rate": 0.00016701349179734816, "grad_norm": 0.5581561923027039, "tokens_per_sec": 134260.33992429238, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2089 |
+
{"event": "train", "epoch": 1, "step": 41740, "loss": 2.0013047754764557, "perplexity": 7.398703450567936, "learning_rate": 0.00016690938469017373, "grad_norm": 0.5684869289398193, "tokens_per_sec": 135325.4407385454, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2090 |
+
{"event": "train", "epoch": 1, "step": 41760, "loss": 2.7114349603652954, "perplexity": 15.050857410437429, "learning_rate": 0.00016680526933222236, "grad_norm": 0.6931224465370178, "tokens_per_sec": 133208.3882689659, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2091 |
+
{"event": "train", "epoch": 1, "step": 41780, "loss": 1.8950715959072113, "perplexity": 6.65302471692197, "learning_rate": 0.00016670114577429625, "grad_norm": 0.5267118811607361, "tokens_per_sec": 133785.53800028813, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2092 |
+
{"event": "train", "epoch": 1, "step": 41800, "loss": 1.8112698793411255, "perplexity": 6.118211892110021, "learning_rate": 0.00016659701406720152, "grad_norm": 0.8399603962898254, "tokens_per_sec": 134490.12303434886, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2093 |
+
{"event": "train", "epoch": 1, "step": 41820, "loss": 2.058238536119461, "perplexity": 7.832161582807077, "learning_rate": 0.0001664928742617482, "grad_norm": 0.5574131011962891, "tokens_per_sec": 135188.20504618605, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2094 |
+
{"event": "train", "epoch": 1, "step": 41840, "loss": 1.7678346633911133, "perplexity": 5.85815474102921, "learning_rate": 0.00016638872640875027, "grad_norm": 0.5610326528549194, "tokens_per_sec": 134233.720689379, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2095 |
+
{"event": "train", "epoch": 1, "step": 41860, "loss": 1.6607941836118698, "perplexity": 5.263489361964729, "learning_rate": 0.00016628457055902578, "grad_norm": 0.5472427606582642, "tokens_per_sec": 134768.7796409734, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2096 |
+
{"event": "train", "epoch": 1, "step": 41880, "loss": 2.152094781398773, "perplexity": 8.602860647666148, "learning_rate": 0.0001661804067633967, "grad_norm": 0.6090496778488159, "tokens_per_sec": 134129.5744808344, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2097 |
+
{"event": "train", "epoch": 1, "step": 41900, "loss": 2.02591472864151, "perplexity": 7.583044200849743, "learning_rate": 0.00016607623507268864, "grad_norm": 0.6402663588523865, "tokens_per_sec": 135198.17885564652, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2098 |
+
{"event": "train", "epoch": 1, "step": 41920, "loss": 1.913996398448944, "perplexity": 6.780130831509509, "learning_rate": 0.00016597205553773139, "grad_norm": 0.5669756531715393, "tokens_per_sec": 130346.0431859911, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2099 |
+
{"event": "train", "epoch": 1, "step": 41940, "loss": 2.5301857590675354, "perplexity": 12.555838280864238, "learning_rate": 0.0001658678682093582, "grad_norm": 0.8043484091758728, "tokens_per_sec": 134328.05015838204, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2100 |
+
{"event": "train", "epoch": 1, "step": 41960, "loss": 2.162017822265625, "perplexity": 8.688652137709427, "learning_rate": 0.00016576367313840643, "grad_norm": 0.5778524279594421, "tokens_per_sec": 134780.27771381725, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2101 |
+
{"event": "train", "epoch": 1, "step": 41980, "loss": 1.9399862885475159, "perplexity": 6.9586555567081225, "learning_rate": 0.00016565947037571694, "grad_norm": 0.5299392938613892, "tokens_per_sec": 134132.71615157972, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2102 |
+
{"event": "train", "epoch": 1, "step": 42000, "loss": 1.8999812006950378, "perplexity": 6.685768753292141, "learning_rate": 0.00016555525997213462, "grad_norm": 0.6322932243347168, "tokens_per_sec": 130927.53440351023, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2103 |
+
{"event": "train", "epoch": 1, "step": 42020, "loss": 2.9777583479881287, "perplexity": 19.643732839706566, "learning_rate": 0.000165451041978508, "grad_norm": 0.699927568435669, "tokens_per_sec": 134870.3477000525, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2104 |
+
{"event": "train", "epoch": 1, "step": 42040, "loss": 1.9582381844520569, "perplexity": 7.086830372019911, "learning_rate": 0.00016534681644568916, "grad_norm": 0.5035626888275146, "tokens_per_sec": 134443.28812597698, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2105 |
+
{"event": "train", "epoch": 1, "step": 42060, "loss": 2.044198215007782, "perplexity": 7.7229638983081, "learning_rate": 0.00016524258342453401, "grad_norm": 0.6404675841331482, "tokens_per_sec": 134741.2979458266, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2106 |
+
{"event": "train", "epoch": 1, "step": 42080, "loss": 2.6453439593315125, "perplexity": 14.088290048668384, "learning_rate": 0.00016513834296590224, "grad_norm": 0.6791557669639587, "tokens_per_sec": 133129.80785146382, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2107 |
+
{"event": "train", "epoch": 1, "step": 42100, "loss": 1.8127930760383606, "perplexity": 6.127538233379985, "learning_rate": 0.00016503409512065676, "grad_norm": 0.5248571634292603, "tokens_per_sec": 134617.90074400857, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2108 |
+
{"event": "train", "epoch": 1, "step": 42120, "loss": 2.2819154858589172, "perplexity": 9.795425449798861, "learning_rate": 0.00016492983993966453, "grad_norm": 0.6088126301765442, "tokens_per_sec": 134016.5722660421, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2109 |
+
{"event": "train", "epoch": 1, "step": 42140, "loss": 2.038443624973297, "perplexity": 7.678649036445057, "learning_rate": 0.00016482557747379581, "grad_norm": 0.5995537638664246, "tokens_per_sec": 135223.18634062255, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2110 |
+
{"event": "train", "epoch": 1, "step": 42160, "loss": 1.8108071386814117, "perplexity": 6.115381401644817, "learning_rate": 0.0001647213077739246, "grad_norm": 0.6026691198348999, "tokens_per_sec": 134808.17199550767, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2111 |
+
{"event": "train", "epoch": 1, "step": 42180, "loss": 1.958040475845337, "perplexity": 7.085429383159339, "learning_rate": 0.00016461703089092804, "grad_norm": 0.5050331950187683, "tokens_per_sec": 134549.76633939418, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2112 |
+
{"event": "train", "epoch": 1, "step": 42200, "loss": 2.073617607355118, "perplexity": 7.95354393524612, "learning_rate": 0.0001645127468756875, "grad_norm": 0.6668148040771484, "tokens_per_sec": 134309.40991164866, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2113 |
+
{"event": "train", "epoch": 1, "step": 42220, "loss": 2.15499547123909, "perplexity": 8.627851105405451, "learning_rate": 0.00016440845577908708, "grad_norm": 0.5562871098518372, "tokens_per_sec": 134345.9076403952, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2114 |
+
{"event": "train", "epoch": 1, "step": 42240, "loss": 1.7796520590782166, "perplexity": 5.927793537783955, "learning_rate": 0.00016430415765201475, "grad_norm": 0.5534659028053284, "tokens_per_sec": 134998.98187361626, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2115 |
+
{"event": "train", "epoch": 1, "step": 42260, "loss": 1.739601969718933, "perplexity": 5.695076158717918, "learning_rate": 0.00016419985254536185, "grad_norm": 0.5022242665290833, "tokens_per_sec": 134068.47208381293, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2116 |
+
{"event": "train", "epoch": 1, "step": 42280, "loss": 2.3809462189674377, "perplexity": 10.815131506467054, "learning_rate": 0.00016409554051002298, "grad_norm": 0.6999700665473938, "tokens_per_sec": 129726.82353940586, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2117 |
+
{"event": "train", "epoch": 1, "step": 42300, "loss": 2.5978129506111145, "perplexity": 13.434324351273894, "learning_rate": 0.0001639912215968963, "grad_norm": 0.7861074805259705, "tokens_per_sec": 127209.29111749548, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2118 |
+
{"event": "train", "epoch": 1, "step": 42320, "loss": 2.5737792551517487, "perplexity": 13.115296951000381, "learning_rate": 0.0001638868958568832, "grad_norm": 1.0568927526474, "tokens_per_sec": 135430.65232861892, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2119 |
+
{"event": "train", "epoch": 1, "step": 42340, "loss": 1.664481669664383, "perplexity": 5.282934234889824, "learning_rate": 0.00016378256334088857, "grad_norm": 0.5237509608268738, "tokens_per_sec": 133685.3365911925, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2120 |
+
{"event": "train", "epoch": 1, "step": 42360, "loss": 2.4947236478328705, "perplexity": 12.118384114091127, "learning_rate": 0.0001636782240998203, "grad_norm": 0.7318628430366516, "tokens_per_sec": 135558.61985597774, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2121 |
+
{"event": "train", "epoch": 1, "step": 42380, "loss": 1.4748049080371857, "perplexity": 4.370183102158597, "learning_rate": 0.0001635738781845899, "grad_norm": 0.7305182218551636, "tokens_per_sec": 134346.03896299395, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2122 |
+
{"event": "train", "epoch": 1, "step": 42400, "loss": 2.1384997963905334, "perplexity": 8.486696301559103, "learning_rate": 0.00016346952564611194, "grad_norm": 0.5524712204933167, "tokens_per_sec": 134593.51203408735, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2123 |
+
{"event": "train", "epoch": 1, "step": 42420, "loss": 2.0804328322410583, "perplexity": 8.007934256416807, "learning_rate": 0.0001633651665353044, "grad_norm": 0.5855236649513245, "tokens_per_sec": 134191.5187190002, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2124 |
+
{"event": "train", "epoch": 1, "step": 42440, "loss": 2.06154265999794, "perplexity": 7.8580828148114055, "learning_rate": 0.00016326080090308808, "grad_norm": 0.5774142146110535, "tokens_per_sec": 133726.70027243716, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2125 |
+
{"event": "train", "epoch": 1, "step": 42460, "loss": 2.1466904282569885, "perplexity": 8.556493156699021, "learning_rate": 0.0001631564288003874, "grad_norm": 0.5544800162315369, "tokens_per_sec": 132478.30340114416, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2126 |
+
{"event": "train", "epoch": 1, "step": 42480, "loss": 2.502295345067978, "perplexity": 12.210489105098864, "learning_rate": 0.00016305205027812968, "grad_norm": 0.6528388857841492, "tokens_per_sec": 135001.89918786075, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2127 |
+
{"event": "train", "epoch": 1, "step": 42500, "loss": 1.899053156375885, "perplexity": 6.679566941804556, "learning_rate": 0.00016294766538724548, "grad_norm": 0.5588608384132385, "tokens_per_sec": 134868.4948290531, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2128 |
+
{"event": "train", "epoch": 1, "step": 42520, "loss": 2.1120263636112213, "perplexity": 8.26497216632237, "learning_rate": 0.00016284327417866854, "grad_norm": 0.5907875299453735, "tokens_per_sec": 135038.24357767607, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2129 |
+
{"event": "train", "epoch": 1, "step": 42540, "loss": 1.8773012459278107, "perplexity": 6.535842408286936, "learning_rate": 0.00016273887670333544, "grad_norm": 0.4967126250267029, "tokens_per_sec": 133587.49247641256, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2130 |
+
{"event": "train", "epoch": 1, "step": 42560, "loss": 1.9364867210388184, "perplexity": 6.934345833386629, "learning_rate": 0.00016263447301218597, "grad_norm": 0.5997911691665649, "tokens_per_sec": 134805.26304783445, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2131 |
+
{"event": "train", "epoch": 1, "step": 42580, "loss": 2.650300621986389, "perplexity": 14.158294300228462, "learning_rate": 0.00016253006315616295, "grad_norm": 0.8494299650192261, "tokens_per_sec": 134939.99961905472, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2132 |
+
{"event": "train", "epoch": 1, "step": 42600, "loss": 2.042030692100525, "perplexity": 7.706242325888941, "learning_rate": 0.0001624256471862122, "grad_norm": 0.5547475814819336, "tokens_per_sec": 134242.50518113206, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2133 |
+
{"event": "train", "epoch": 1, "step": 42620, "loss": 2.1832743883132935, "perplexity": 8.87531997571535, "learning_rate": 0.0001623212251532826, "grad_norm": 0.5718026757240295, "tokens_per_sec": 133452.46154085168, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2134 |
+
{"event": "train", "epoch": 1, "step": 42640, "loss": 2.2155624330043793, "perplexity": 9.166563239749815, "learning_rate": 0.00016221679710832574, "grad_norm": 0.6618753671646118, "tokens_per_sec": 133552.05447845894, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2135 |
+
{"event": "train", "epoch": 1, "step": 42660, "loss": 2.3206330835819244, "perplexity": 10.182118398020696, "learning_rate": 0.00016211236310229648, "grad_norm": 0.8416472673416138, "tokens_per_sec": 134369.5498448931, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2136 |
+
{"event": "train", "epoch": 1, "step": 42680, "loss": 2.1881091594696045, "perplexity": 8.91833401446613, "learning_rate": 0.00016200792318615232, "grad_norm": 0.5740881562232971, "tokens_per_sec": 134770.1011587532, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2137 |
+
{"event": "train", "epoch": 1, "step": 42700, "loss": 2.0365380346775055, "perplexity": 7.664030610145692, "learning_rate": 0.00016190347741085385, "grad_norm": 0.5252773761749268, "tokens_per_sec": 134387.8125752906, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2138 |
+
{"event": "train", "epoch": 1, "step": 42720, "loss": 1.958022564649582, "perplexity": 7.085302475783184, "learning_rate": 0.00016179902582736432, "grad_norm": 0.6194599270820618, "tokens_per_sec": 134454.33611883412, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2139 |
+
{"event": "train", "epoch": 1, "step": 42740, "loss": 2.8357678949832916, "perplexity": 17.043482881731745, "learning_rate": 0.00016169456848664994, "grad_norm": 1.0921558141708374, "tokens_per_sec": 134306.25995846873, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2140 |
+
{"event": "train", "epoch": 1, "step": 42760, "loss": 1.9089230000972748, "perplexity": 6.745819637697571, "learning_rate": 0.00016159010543967967, "grad_norm": 0.5326470732688904, "tokens_per_sec": 133773.29757193386, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2141 |
+
{"event": "train", "epoch": 1, "step": 42780, "loss": 2.7193676233291626, "perplexity": 15.170725597978034, "learning_rate": 0.0001614856367374252, "grad_norm": 0.715214729309082, "tokens_per_sec": 134609.198892874, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2142 |
+
{"event": "train", "epoch": 1, "step": 42800, "loss": 1.9626305401325226, "perplexity": 7.118026714221035, "learning_rate": 0.00016138116243086134, "grad_norm": 0.5575718879699707, "tokens_per_sec": 133643.3487248651, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2143 |
+
{"event": "train", "epoch": 1, "step": 42820, "loss": 2.310196727514267, "perplexity": 10.076406766495177, "learning_rate": 0.00016127668257096513, "grad_norm": 0.7226421236991882, "tokens_per_sec": 133986.00025931714, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2144 |
+
{"event": "train", "epoch": 1, "step": 42840, "loss": 1.6900624632835388, "perplexity": 5.4198192342638345, "learning_rate": 0.00016117219720871651, "grad_norm": 0.4762110710144043, "tokens_per_sec": 134739.18443350584, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2145 |
+
{"event": "train", "epoch": 1, "step": 42860, "loss": 3.129296600818634, "perplexity": 22.857895661313638, "learning_rate": 0.0001610677063950983, "grad_norm": 0.9525426626205444, "tokens_per_sec": 132202.9244463768, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2146 |
+
{"event": "train", "epoch": 1, "step": 42880, "loss": 1.8932980000972748, "perplexity": 6.641235397994374, "learning_rate": 0.0001609632101810957, "grad_norm": 0.5573139786720276, "tokens_per_sec": 134214.05808434926, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|
| 2147 |
+
{"event": "train", "epoch": 1, "step": 42900, "loss": 2.8735920190811157, "perplexity": 17.70048462380793, "learning_rate": 0.00016085870861769677, "grad_norm": 0.8360785841941833, "tokens_per_sec": 131752.6412291894, "gpu_mem_alloc_gb": 4.000064849853516, "gpu_mem_reserved_gb": 53.087890625, "grad_scale": 1.0}
|