Upload folder using huggingface_hub

newton/README.md

@@ -1,9 +1,9 @@
 ---
-base_model: meta-llama/Llama-3.1-8B-Instruct
+base_model: Raiff1982/codette-llama-3.1-8b-merged
 library_name: peft
 model_name: newton
 tags:
-- base_model:adapter:meta-llama/Llama-3.1-8B-Instruct
+- base_model:adapter:Raiff1982/codette-llama-3.1-8b-merged
 - lora
 - sft
 - transformers
@@ -14,7 +14,7 @@ pipeline_tag: text-generation
 
 # Model Card for newton
 
-This model is a fine-tuned version of [meta-llama/Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-8B-Instruct).
+This model is a fine-tuned version of [Raiff1982/codette-llama-3.1-8b-merged](https://huggingface.co/Raiff1982/codette-llama-3.1-8b-merged).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -39,10 +39,10 @@ This model was trained with SFT.
 ### Framework versions
 
 - PEFT 0.18.1
-- TRL: 0.29.1
-- Transformers: 5.3.0
-- Pytorch: 2.10.0
-- Datasets: 4.8.3
+- TRL: 1.0.0
+- Transformers: 5.5.0
+- Pytorch: 2.11.0
+- Datasets: 4.8.4
 - Tokenizers: 0.22.2
 
 ## Citations

newton/adapter_config.json

@@ -3,7 +3,7 @@
   "alpha_pattern": {},
   "arrow_config": null,
   "auto_mapping": null,
-  "base_model_name_or_path": "meta-llama/Llama-3.1-8B-Instruct",
+  "base_model_name_or_path": "Raiff1982/codette-llama-3.1-8b-merged",
   "bias": "none",
   "corda_config": null,
   "ensure_weight_tying": false,
@@ -29,10 +29,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "v_proj",
     "q_proj",
-    "k_proj"
+    "k_proj",
+    "o_proj",
+    "v_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

newton/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15e7ab5a4223f583a958b59b89be7c7a7a2892c32c63e7d8880e92c28582f30e
+oid sha256:e66a75cae8fde6774170b0e340f267894de55426775f0c1693a12ce630003af8
 size 27297544

newton/checkpoint-309/README.md

@@ -1,9 +1,9 @@
 ---
-base_model: meta-llama/Llama-3.1-8B-Instruct
+base_model: Raiff1982/codette-llama-3.1-8b-merged
 library_name: peft
 pipeline_tag: text-generation
 tags:
-- base_model:adapter:meta-llama/Llama-3.1-8B-Instruct
+- base_model:adapter:Raiff1982/codette-llama-3.1-8b-merged
 - lora
 - sft
 - transformers

newton/checkpoint-309/adapter_config.json

@@ -3,7 +3,7 @@
   "alpha_pattern": {},
   "arrow_config": null,
   "auto_mapping": null,
-  "base_model_name_or_path": "meta-llama/Llama-3.1-8B-Instruct",
+  "base_model_name_or_path": "Raiff1982/codette-llama-3.1-8b-merged",
   "bias": "none",
   "corda_config": null,
   "ensure_weight_tying": false,
@@ -29,10 +29,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "v_proj",
     "q_proj",
-    "k_proj"
+    "k_proj",
+    "o_proj",
+    "v_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

newton/checkpoint-309/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15e7ab5a4223f583a958b59b89be7c7a7a2892c32c63e7d8880e92c28582f30e
+oid sha256:e66a75cae8fde6774170b0e340f267894de55426775f0c1693a12ce630003af8
 size 27297544

newton/checkpoint-309/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c0048ede16e73e75ae0558a8be34314fdf4dd7a0048a2c0345323e009f8cd75a
+oid sha256:a9cc12bcba29c33eaebecec0edb4b744ab01f0d521051ab6ce9ac51ff823612a
 size 54745547

newton/checkpoint-309/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a3ce3a6237917355c7ff6bb395ee6541fd028f6e898cd29ab73f2c1037e36ca
+oid sha256:167c516cb3da9c574bcafa89ab7815d70767a2593edafa4119230d61545ccb9f
 size 14645

newton/checkpoint-309/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dfae2b0e67a84fe0276ee2a4bd443baa818fcadf7c94eb154f2c9f40764c3223
+oid sha256:b805c703bca5d31d07f2c000dd76db9e70297bbf6528c7f0fe3645a9d4d5a44c
 size 1465

newton/checkpoint-309/trainer_state.json

@@ -10,303 +10,303 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "entropy": 2.5801601350307464,
-      "epoch": 0.0975609756097561,
-      "grad_norm": 0.640625,
-      "learning_rate": 0.00018,
-      "loss": 3.2993820190429686,
-      "mean_token_accuracy": 0.4129736639559269,
-      "num_tokens": 20093.0,
+      "entropy": 2.487887626886368,
+      "epoch": 0.0970873786407767,
+      "grad_norm": 0.55078125,
+      "learning_rate": 9e-05,
+      "loss": 3.4457256317138674,
+      "mean_token_accuracy": 0.4037958011031151,
+      "num_tokens": 20045.0,
       "step": 10
     },
     {
-      "entropy": 1.46146529763937,
-      "epoch": 0.1951219512195122,
-      "grad_norm": 0.7109375,
-      "learning_rate": 0.0001939799331103679,
-      "loss": 1.3410646438598632,
-      "mean_token_accuracy": 0.6868809185922146,
-      "num_tokens": 40396.0,
+      "entropy": 2.367420971393585,
+      "epoch": 0.1941747572815534,
+      "grad_norm": 1.03125,
+      "learning_rate": 9.698996655518396e-05,
+      "loss": 2.301749038696289,
+      "mean_token_accuracy": 0.5057214729487896,
+      "num_tokens": 40110.0,
       "step": 20
     },
     {
-      "entropy": 0.30421045832335947,
-      "epoch": 0.2926829268292683,
-      "grad_norm": 0.1591796875,
-      "learning_rate": 0.00018729096989966558,
-      "loss": 0.2733942985534668,
-      "mean_token_accuracy": 0.9430402711033821,
-      "num_tokens": 60559.0,
+      "entropy": 0.9785687401890755,
+      "epoch": 0.2912621359223301,
+      "grad_norm": 0.482421875,
+      "learning_rate": 9.364548494983279e-05,
+      "loss": 0.8281048774719239,
+      "mean_token_accuracy": 0.8013610184192658,
+      "num_tokens": 60360.0,
       "step": 30
     },
     {
-      "entropy": 0.1860255692154169,
-      "epoch": 0.3902439024390244,
-      "grad_norm": 0.11865234375,
-      "learning_rate": 0.00018060200668896322,
-      "loss": 0.17133373022079468,
-      "mean_token_accuracy": 0.9654295533895493,
-      "num_tokens": 80726.0,
+      "entropy": 0.3154874790459871,
+      "epoch": 0.3883495145631068,
+      "grad_norm": 0.275390625,
+      "learning_rate": 9.030100334448161e-05,
+      "loss": 0.3118274450302124,
+      "mean_token_accuracy": 0.936956076323986,
+      "num_tokens": 80424.0,
       "step": 40
     },
     {
-      "entropy": 0.16763325594365597,
-      "epoch": 0.4878048780487805,
-      "grad_norm": 0.11328125,
-      "learning_rate": 0.00017391304347826088,
-      "loss": 0.1414021611213684,
-      "mean_token_accuracy": 0.9735398545861245,
-      "num_tokens": 100934.0,
+      "entropy": 0.21394508741796017,
+      "epoch": 0.4854368932038835,
+      "grad_norm": 0.2236328125,
+      "learning_rate": 8.695652173913044e-05,
+      "loss": 0.1942020535469055,
+      "mean_token_accuracy": 0.9629994183778763,
+      "num_tokens": 100413.0,
       "step": 50
     },
     {
-      "entropy": 0.14396543558686972,
-      "epoch": 0.5853658536585366,
-      "grad_norm": 0.1025390625,
-      "learning_rate": 0.00016722408026755855,
-      "loss": 0.11398149728775024,
-      "mean_token_accuracy": 0.9769163399934768,
-      "num_tokens": 121033.0,
+      "entropy": 0.17603826858103275,
+      "epoch": 0.5825242718446602,
+      "grad_norm": 0.162109375,
+      "learning_rate": 8.361204013377927e-05,
+      "loss": 0.17068777084350586,
+      "mean_token_accuracy": 0.9660190954804421,
+      "num_tokens": 120668.0,
       "step": 60
     },
     {
-      "entropy": 0.09721304103732109,
-      "epoch": 0.6829268292682927,
-      "grad_norm": 0.1484375,
-      "learning_rate": 0.00016053511705685619,
-      "loss": 0.09096546769142151,
-      "mean_token_accuracy": 0.9776118725538254,
-      "num_tokens": 141112.0,
+      "entropy": 0.1720115825533867,
+      "epoch": 0.6796116504854369,
+      "grad_norm": 0.1533203125,
+      "learning_rate": 8.026755852842809e-05,
+      "loss": 0.15503238439559935,
+      "mean_token_accuracy": 0.9691305428743362,
+      "num_tokens": 140907.0,
       "step": 70
     },
     {
-      "entropy": 0.10199905186891556,
-      "epoch": 0.7804878048780488,
-      "grad_norm": 0.09765625,
-      "learning_rate": 0.00015384615384615385,
-      "loss": 0.07685146927833557,
-      "mean_token_accuracy": 0.9789551600813866,
-      "num_tokens": 161349.0,
+      "entropy": 0.15427112840116025,
+      "epoch": 0.7766990291262136,
+      "grad_norm": 0.177734375,
+      "learning_rate": 7.692307692307693e-05,
+      "loss": 0.14039652347564696,
+      "mean_token_accuracy": 0.9698389366269111,
+      "num_tokens": 161024.0,
       "step": 80
     },
     {
-      "entropy": 0.09305545147508383,
-      "epoch": 0.8780487804878049,
-      "grad_norm": 0.0869140625,
-      "learning_rate": 0.0001471571906354515,
-      "loss": 0.06557443141937255,
-      "mean_token_accuracy": 0.9781694248318672,
-      "num_tokens": 181560.0,
+      "entropy": 0.15376258343458177,
+      "epoch": 0.8737864077669902,
+      "grad_norm": 0.130859375,
+      "learning_rate": 7.357859531772575e-05,
+      "loss": 0.1298056960105896,
+      "mean_token_accuracy": 0.9742556199431419,
+      "num_tokens": 181243.0,
       "step": 90
     },
     {
-      "entropy": 0.0788642093539238,
-      "epoch": 0.975609756097561,
-      "grad_norm": 0.051025390625,
-      "learning_rate": 0.00014046822742474916,
-      "loss": 0.05446377396583557,
-      "mean_token_accuracy": 0.9795627012848854,
-      "num_tokens": 201709.0,
+      "entropy": 0.14775074161589147,
+      "epoch": 0.970873786407767,
+      "grad_norm": 0.126953125,
+      "learning_rate": 7.023411371237458e-05,
+      "loss": 0.12051811218261718,
+      "mean_token_accuracy": 0.9776876136660576,
+      "num_tokens": 201412.0,
       "step": 100
     },
     {
-      "entropy": 0.06586153550367606,
-      "epoch": 1.0682926829268293,
-      "grad_norm": 0.056640625,
-      "learning_rate": 0.00013377926421404682,
-      "loss": 0.05286722779273987,
-      "mean_token_accuracy": 0.9784782999440244,
-      "num_tokens": 220911.0,
+      "entropy": 0.15358976498246193,
+      "epoch": 1.0679611650485437,
+      "grad_norm": 0.1357421875,
+      "learning_rate": 6.688963210702341e-05,
+      "loss": 0.13191524744033814,
+      "mean_token_accuracy": 0.9737549498677254,
+      "num_tokens": 221317.0,
       "step": 110
     },
     {
-      "entropy": 0.06249262308701873,
-      "epoch": 1.1658536585365853,
-      "grad_norm": 0.0498046875,
-      "learning_rate": 0.0001270903010033445,
-      "loss": 0.052178531885147095,
-      "mean_token_accuracy": 0.9783070877194404,
-      "num_tokens": 241100.0,
+      "entropy": 0.12130333054810763,
+      "epoch": 1.1650485436893203,
+      "grad_norm": 0.111328125,
+      "learning_rate": 6.354515050167224e-05,
+      "loss": 0.09631805419921875,
+      "mean_token_accuracy": 0.9778564915060997,
+      "num_tokens": 241469.0,
       "step": 120
     },
     {
-      "entropy": 0.05583516648039222,
-      "epoch": 1.2634146341463415,
-      "grad_norm": 0.0576171875,
-      "learning_rate": 0.00012040133779264215,
-      "loss": 0.04987884163856506,
-      "mean_token_accuracy": 0.9790923863649368,
-      "num_tokens": 261282.0,
+      "entropy": 0.12368765268474817,
+      "epoch": 1.262135922330097,
+      "grad_norm": 0.126953125,
+      "learning_rate": 6.0200668896321076e-05,
+      "loss": 0.10474776029586792,
+      "mean_token_accuracy": 0.9744847908616066,
+      "num_tokens": 261431.0,
       "step": 130
     },
     {
-      "entropy": 0.05573393665254116,
-      "epoch": 1.3609756097560974,
-      "grad_norm": 0.05078125,
-      "learning_rate": 0.00011371237458193979,
-      "loss": 0.049983879923820494,
-      "mean_token_accuracy": 0.9795695841312408,
-      "num_tokens": 281470.0,
+      "entropy": 0.10261523667722941,
+      "epoch": 1.3592233009708738,
+      "grad_norm": 0.11572265625,
+      "learning_rate": 5.6856187290969896e-05,
+      "loss": 0.076785808801651,
+      "mean_token_accuracy": 0.9786087408661842,
+      "num_tokens": 281589.0,
       "step": 140
     },
     {
-      "entropy": 0.05604505110532045,
-      "epoch": 1.4585365853658536,
-      "grad_norm": 0.049560546875,
-      "learning_rate": 0.00010702341137123746,
-      "loss": 0.049191102385520935,
-      "mean_token_accuracy": 0.9789946660399437,
-      "num_tokens": 301537.0,
+      "entropy": 0.09858085233718157,
+      "epoch": 1.4563106796116505,
+      "grad_norm": 0.10693359375,
+      "learning_rate": 5.351170568561873e-05,
+      "loss": 0.07057241201400757,
+      "mean_token_accuracy": 0.978242176771164,
+      "num_tokens": 301814.0,
       "step": 150
     },
     {
-      "entropy": 0.05383661286905408,
-      "epoch": 1.5560975609756098,
-      "grad_norm": 0.052001953125,
-      "learning_rate": 0.00010033444816053512,
-      "loss": 0.04975903928279877,
-      "mean_token_accuracy": 0.9779537573456765,
-      "num_tokens": 321687.0,
+      "entropy": 0.09467246811836957,
+      "epoch": 1.5533980582524272,
+      "grad_norm": 0.1171875,
+      "learning_rate": 5.016722408026756e-05,
+      "loss": 0.06553536057472228,
+      "mean_token_accuracy": 0.9788320794701576,
+      "num_tokens": 322034.0,
       "step": 160
     },
     {
-      "entropy": 0.05550122009590268,
-      "epoch": 1.653658536585366,
-      "grad_norm": 0.05517578125,
-      "learning_rate": 9.364548494983279e-05,
-      "loss": 0.04978601038455963,
-      "mean_token_accuracy": 0.9779243767261505,
-      "num_tokens": 341947.0,
+      "entropy": 0.09434448610991239,
+      "epoch": 1.650485436893204,
+      "grad_norm": 0.07666015625,
+      "learning_rate": 4.6822742474916394e-05,
+      "loss": 0.06265009045600892,
+      "mean_token_accuracy": 0.978448674082756,
+      "num_tokens": 342246.0,
       "step": 170
     },
     {
-      "entropy": 0.0521472885273397,
-      "epoch": 1.751219512195122,
-      "grad_norm": 0.053466796875,
-      "learning_rate": 8.695652173913044e-05,
-      "loss": 0.04891734123229981,
-      "mean_token_accuracy": 0.9793664410710334,
-      "num_tokens": 362157.0,
+      "entropy": 0.090944261290133,
+      "epoch": 1.7475728155339807,
+      "grad_norm": 0.10888671875,
+      "learning_rate": 4.347826086956522e-05,
+      "loss": 0.05846574306488037,
+      "mean_token_accuracy": 0.9787591323256493,
+      "num_tokens": 362493.0,
       "step": 180
     },
     {
-      "entropy": 0.05427801813930273,
-      "epoch": 1.848780487804878,
-      "grad_norm": 0.03564453125,
-      "learning_rate": 8.026755852842809e-05,
-      "loss": 0.04973878860473633,
-      "mean_token_accuracy": 0.9789470374584198,
-      "num_tokens": 382284.0,
+      "entropy": 0.10327567681670188,
+      "epoch": 1.8446601941747574,
+      "grad_norm": 0.09619140625,
+      "learning_rate": 4.0133779264214046e-05,
+      "loss": 0.07640682458877564,
+      "mean_token_accuracy": 0.9750214621424675,
+      "num_tokens": 382470.0,
       "step": 190
     },
     {
-      "entropy": 0.05222779270261526,
-      "epoch": 1.946341463414634,
-      "grad_norm": 0.056884765625,
-      "learning_rate": 7.357859531772575e-05,
-      "loss": 0.04821131825447082,
-      "mean_token_accuracy": 0.9792478010058403,
-      "num_tokens": 402458.0,
+      "entropy": 0.08134665545076132,
+      "epoch": 1.941747572815534,
+      "grad_norm": 0.08447265625,
+      "learning_rate": 3.678929765886287e-05,
+      "loss": 0.053827738761901854,
+      "mean_token_accuracy": 0.9790568739175797,
+      "num_tokens": 402761.0,
       "step": 200
     },
     {
-      "entropy": 0.0516971405595541,
-      "epoch": 2.0390243902439025,
-      "grad_norm": 0.04345703125,
-      "learning_rate": 6.688963210702341e-05,
-      "loss": 0.04751978814601898,
-      "mean_token_accuracy": 0.9792861828678533,
-      "num_tokens": 421706.0,
+      "entropy": 0.09850292447954416,
+      "epoch": 2.0388349514563107,
+      "grad_norm": 0.1845703125,
+      "learning_rate": 3.3444816053511705e-05,
+      "loss": 0.0715977430343628,
+      "mean_token_accuracy": 0.9744585126638412,
+      "num_tokens": 422580.0,
       "step": 210
     },
     {
-      "entropy": 0.05148327555507422,
-      "epoch": 2.1365853658536587,
-      "grad_norm": 0.043701171875,
-      "learning_rate": 6.0200668896321076e-05,
-      "loss": 0.04758384227752686,
-      "mean_token_accuracy": 0.9795778721570969,
-      "num_tokens": 442021.0,
+      "entropy": 0.07439298946410418,
+      "epoch": 2.1359223300970873,
+      "grad_norm": 0.07763671875,
+      "learning_rate": 3.0100334448160538e-05,
+      "loss": 0.05062834024429321,
+      "mean_token_accuracy": 0.9796976149082184,
+      "num_tokens": 442815.0,
       "step": 220
     },
     {
-      "entropy": 0.05236647073179483,
-      "epoch": 2.234146341463415,
-      "grad_norm": 0.040283203125,
-      "learning_rate": 5.351170568561873e-05,
-      "loss": 0.04777625799179077,
-      "mean_token_accuracy": 0.9786792114377022,
-      "num_tokens": 462138.0,
+      "entropy": 0.0704705884680152,
+      "epoch": 2.233009708737864,
+      "grad_norm": 0.09716796875,
+      "learning_rate": 2.6755852842809364e-05,
+      "loss": 0.0514835000038147,
+      "mean_token_accuracy": 0.9794572740793228,
+      "num_tokens": 462923.0,
       "step": 230
     },
     {
-      "entropy": 0.05254681948572397,
-      "epoch": 2.3317073170731706,
-      "grad_norm": 0.03857421875,
-      "learning_rate": 4.6822742474916394e-05,
-      "loss": 0.047863802313804625,
-      "mean_token_accuracy": 0.9788415163755417,
-      "num_tokens": 482318.0,
+      "entropy": 0.07191200498491526,
+      "epoch": 2.3300970873786406,
+      "grad_norm": 0.07958984375,
+      "learning_rate": 2.3411371237458197e-05,
+      "loss": 0.051685428619384764,
+      "mean_token_accuracy": 0.9794511407613754,
+      "num_tokens": 483078.0,
       "step": 240
     },
     {
-      "entropy": 0.0514264177531004,
-      "epoch": 2.4292682926829268,
-      "grad_norm": 0.041259765625,
-      "learning_rate": 4.0133779264214046e-05,
-      "loss": 0.04747449159622193,
-      "mean_token_accuracy": 0.9798328444361687,
-      "num_tokens": 502473.0,
+      "entropy": 0.06878451202064753,
+      "epoch": 2.4271844660194173,
+      "grad_norm": 0.07470703125,
+      "learning_rate": 2.0066889632107023e-05,
+      "loss": 0.050105297565460206,
+      "mean_token_accuracy": 0.9797055780887604,
+      "num_tokens": 503252.0,
       "step": 250
     },
     {
-      "entropy": 0.05115404771640897,
-      "epoch": 2.526829268292683,
-      "grad_norm": 0.0576171875,
-      "learning_rate": 3.3444816053511705e-05,
-      "loss": 0.04702305793762207,
-      "mean_token_accuracy": 0.9793446376919747,
-      "num_tokens": 522684.0,
+      "entropy": 0.08662745058536529,
+      "epoch": 2.524271844660194,
+      "grad_norm": 0.08837890625,
+      "learning_rate": 1.6722408026755853e-05,
+      "loss": 0.06598815321922302,
+      "mean_token_accuracy": 0.9757005825638772,
+      "num_tokens": 523400.0,
       "step": 260
     },
     {
-      "entropy": 0.05094934357330203,
-      "epoch": 2.624390243902439,
-      "grad_norm": 0.052734375,
-      "learning_rate": 2.6755852842809364e-05,
-      "loss": 0.04771294593811035,
-      "mean_token_accuracy": 0.9783476650714874,
-      "num_tokens": 542810.0,
+      "entropy": 0.08474288573488593,
+      "epoch": 2.6213592233009706,
+      "grad_norm": 0.0791015625,
+      "learning_rate": 1.3377926421404682e-05,
+      "loss": 0.0655213713645935,
+      "mean_token_accuracy": 0.976191833615303,
+      "num_tokens": 543416.0,
       "step": 270
     },
     {
-      "entropy": 0.0513969948515296,
-      "epoch": 2.721951219512195,
-      "grad_norm": 0.0517578125,
-      "learning_rate": 2.0066889632107023e-05,
-      "loss": 0.04680090546607971,
-      "mean_token_accuracy": 0.9800771608948707,
-      "num_tokens": 563051.0,
+      "entropy": 0.06866553500294685,
+      "epoch": 2.7184466019417477,
+      "grad_norm": 0.072265625,
+      "learning_rate": 1.0033444816053512e-05,
+      "loss": 0.04991424381732941,
+      "mean_token_accuracy": 0.9802871778607368,
+      "num_tokens": 563552.0,
       "step": 280
     },
     {
-      "entropy": 0.05136826252564788,
-      "epoch": 2.819512195121951,
-      "grad_norm": 0.0400390625,
-      "learning_rate": 1.3377926421404682e-05,
-      "loss": 0.0476355105638504,
-      "mean_token_accuracy": 0.9787444621324539,
-      "num_tokens": 583158.0,
+      "entropy": 0.06758161811158062,
+      "epoch": 2.8155339805825244,
+      "grad_norm": 0.07373046875,
+      "learning_rate": 6.688963210702341e-06,
+      "loss": 0.049510461091995236,
+      "mean_token_accuracy": 0.9791449457406998,
+      "num_tokens": 583870.0,
       "step": 290
     },
     {
-      "entropy": 0.05182668277993798,
-      "epoch": 2.9170731707317072,
-      "grad_norm": 0.045654296875,
-      "learning_rate": 6.688963210702341e-06,
-      "loss": 0.04734614789485932,
-      "mean_token_accuracy": 0.978977257013321,
-      "num_tokens": 603264.0,
+      "entropy": 0.06784124514088034,
+      "epoch": 2.912621359223301,
+      "grad_norm": 0.12890625,
+      "learning_rate": 3.3444816053511705e-06,
+      "loss": 0.04954342842102051,
+      "mean_token_accuracy": 0.9807449117302894,
+      "num_tokens": 603997.0,
       "step": 300
     }
   ],
@@ -327,7 +327,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2.857929508159488e+16,
+  "total_flos": 2.867826935488512e+16,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

newton/checkpoint-309/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8755273dccefb3d7fa41448d64a8c28d76451700a997d4cbd5f7ac202a091f77
-size 5585
+oid sha256:acff9661de685950df198faf07a4951237f5448c33d774ab0fc779a61a3dd677
+size 5649