Upload 19 files

latest

@@ -1 +1 @@
-global_step100
+global_step75

rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36d2a2034ebb05cb71c510897f2795b31164e50f17b270bc25d2be3ad9a17b22
+oid sha256:08282b46825aa78d10fe10e3fea89555c5b5a691b261a3ddfd58fcb58370edff
 size 15984

rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:060dfdb1c49102cbdc8868a6031e68787601b4ccd782f3fb9b137e20c1fd2c7a
+oid sha256:dbab71d98a3a9a92df82a6bba463947327c3a1bcf35cd9f4f46114641fc42dd9
 size 15984

rng_state_2.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af01895cb66e616591f2e4baa8dcd8151530eab133c73571ccb31c74f35422ce
+oid sha256:caac82d57d878d30219a4f9ec289a97ff90c53afc160b968f251b3fd3454b8d8
 size 15984

rng_state_3.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:677921992b1e0cef3aee776f245975003d22f51d9bd6ed20f248ded1deb72fa9
+oid sha256:19762d2d370222b01817da11bbaa6665d542293373186d66f754e7246bb861ed
 size 15984

rng_state_4.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d69353c629541c690c5471f8ec05fdab2bfecf3d37afaa436bc45939da6db68f
+oid sha256:00c7508b346a7d3c5c23392845f1d013331114ade778794b76e919cb3ed5d33e
 size 15984

rng_state_5.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e40ba6668cc03c9162c68a933d164bf38ae2d196a9a6fec03ae615491201185
+oid sha256:b89de7d14dd20a191f56b74c816ef8b7fe5c171e31efbeadbf321c4539ed68c3
 size 15984

rng_state_6.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:870968fea834e24b2e099cf3e4fe1e3fb8caf38d8f8e5b790d7d47386d4d05f5
+oid sha256:1c71152053553e6e22d670fbc4fd7550bf8a046b54cad7b71869787986a6a42c
 size 15984

rng_state_7.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9e19618bee7c6ef43256fea25abe19bca88535eb1e7dc213cde8929ae4e8180
+oid sha256:4b67db12a26a26ffe03d9afc84a43857eb2e5b2fec2dd189653b415f74208190
 size 15984

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:864006527e551a970ff14e22c9bcc63fb4197ebca7f1b26b2aceedaf636346f3
+oid sha256:2b1a13f720f74c7a907e66559bf8eb2582d7eef15e971987100e6a23b1264031
 size 1064

trainer_state.json

@@ -1,320 +1,242 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.125,
+  "epoch": 2.34375,
   "eval_steps": 500,
-  "global_step": 100,
+  "global_step": 75,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.06,
-      "learning_rate": 4.9980725906018074e-05,
-      "loss": 0.7547,
+      "learning_rate": 4.996988640512931e-05,
+      "loss": 0.7489,
       "step": 2
     },
     {
       "epoch": 0.12,
-      "learning_rate": 4.99229333433282e-05,
-      "loss": 0.566,
+      "learning_rate": 4.987961816680492e-05,
+      "loss": 0.5822,
       "step": 4
     },
     {
       "epoch": 0.19,
-      "learning_rate": 4.982671142387316e-05,
-      "loss": 0.5632,
+      "learning_rate": 4.972941274911953e-05,
+      "loss": 0.5713,
       "step": 6
     },
     {
       "epoch": 0.25,
-      "learning_rate": 4.9692208514878444e-05,
-      "loss": 0.5385,
+      "learning_rate": 4.951963201008076e-05,
+      "loss": 0.5457,
       "step": 8
     },
     {
       "epoch": 0.31,
-      "learning_rate": 4.951963201008076e-05,
-      "loss": 0.5242,
+      "learning_rate": 4.9250781329863606e-05,
+      "loss": 0.5473,
       "step": 10
     },
     {
       "epoch": 0.38,
-      "learning_rate": 4.9309248009941914e-05,
-      "loss": 0.4853,
+      "learning_rate": 4.892350839330522e-05,
+      "loss": 0.5339,
       "step": 12
     },
     {
       "epoch": 0.44,
-      "learning_rate": 4.906138091134118e-05,
-      "loss": 0.5078,
+      "learning_rate": 4.853860162957552e-05,
+      "loss": 0.5368,
       "step": 14
     },
     {
       "epoch": 0.5,
-      "learning_rate": 4.877641290737884e-05,
-      "loss": 0.519,
+      "learning_rate": 4.8096988312782174e-05,
+      "loss": 0.5598,
       "step": 16
     },
     {
       "epoch": 0.56,
-      "learning_rate": 4.8454783398062106e-05,
-      "loss": 0.5759,
+      "learning_rate": 4.759973232808609e-05,
+      "loss": 0.517,
       "step": 18
     },
     {
       "epoch": 0.62,
-      "learning_rate": 4.8096988312782174e-05,
-      "loss": 0.5073,
+      "learning_rate": 4.7048031608708876e-05,
+      "loss": 0.5279,
       "step": 20
     },
     {
       "epoch": 0.69,
-      "learning_rate": 4.7703579345627035e-05,
-      "loss": 0.4996,
+      "learning_rate": 4.6443215250006806e-05,
+      "loss": 0.5079,
       "step": 22
     },
     {
       "epoch": 0.75,
-      "learning_rate": 4.72751631047092e-05,
-      "loss": 0.498,
+      "learning_rate": 4.5786740307563636e-05,
+      "loss": 0.5045,
       "step": 24
     },
     {
       "epoch": 0.81,
-      "learning_rate": 4.681240017681993e-05,
-      "loss": 0.5313,
+      "learning_rate": 4.508018828701612e-05,
+      "loss": 0.5158,
       "step": 26
     },
     {
       "epoch": 0.88,
-      "learning_rate": 4.6316004108852305e-05,
-      "loss": 0.51,
+      "learning_rate": 4.4325261334068426e-05,
+      "loss": 0.5064,
       "step": 28
     },
     {
       "epoch": 0.94,
-      "learning_rate": 4.5786740307563636e-05,
-      "loss": 0.5182,
+      "learning_rate": 4.352377813387398e-05,
+      "loss": 0.5072,
       "step": 30
     },
     {
       "epoch": 1.0,
-      "learning_rate": 4.522542485937369e-05,
-      "loss": 0.5222,
+      "learning_rate": 4.267766952966369e-05,
+      "loss": 0.5087,
       "step": 32
     },
     {
       "epoch": 1.06,
-      "learning_rate": 4.463292327201862e-05,
-      "loss": 0.3137,
+      "learning_rate": 4.178897387117546e-05,
+      "loss": 0.3132,
       "step": 34
     },
     {
       "epoch": 1.12,
-      "learning_rate": 4.401014914000078e-05,
-      "loss": 0.2685,
+      "learning_rate": 4.085983210409114e-05,
+      "loss": 0.2972,
       "step": 36
     },
     {
       "epoch": 1.19,
-      "learning_rate": 4.335806273589214e-05,
-      "loss": 0.2834,
+      "learning_rate": 3.9892482612310836e-05,
+      "loss": 0.2747,
       "step": 38
     },
     {
       "epoch": 1.25,
-      "learning_rate": 4.267766952966369e-05,
-      "loss": 0.2713,
+      "learning_rate": 3.888925582549006e-05,
+      "loss": 0.2399,
       "step": 40
     },
     {
       "epoch": 1.31,
-      "learning_rate": 4.197001863832355e-05,
-      "loss": 0.2705,
+      "learning_rate": 3.785256860483054e-05,
+      "loss": 0.2741,
       "step": 42
     },
     {
       "epoch": 1.38,
-      "learning_rate": 4.123620120825459e-05,
-      "loss": 0.2517,
+      "learning_rate": 3.678491842064995e-05,
+      "loss": 0.2622,
       "step": 44
     },
     {
       "epoch": 1.44,
-      "learning_rate": 4.047734873274586e-05,
-      "loss": 0.2626,
+      "learning_rate": 3.624028324136517e-05,
+      "loss": 0.258,
       "step": 46
     },
     {
       "epoch": 1.5,
-      "learning_rate": 3.969463130731183e-05,
-      "loss": 0.2754,
+      "learning_rate": 3.513103285012475e-05,
+      "loss": 0.2574,
       "step": 48
     },
     {
       "epoch": 1.56,
-      "learning_rate": 3.888925582549006e-05,
-      "loss": 0.2645,
+      "learning_rate": 3.399737591337471e-05,
+      "loss": 0.2551,
       "step": 50
     },
     {
       "epoch": 1.62,
-      "learning_rate": 3.8062464117898724e-05,
-      "loss": 0.2405,
+      "learning_rate": 3.284204350997229e-05,
+      "loss": 0.265,
       "step": 52
     },
     {
       "epoch": 1.69,
-      "learning_rate": 3.721553103742388e-05,
-      "loss": 0.2412,
+      "learning_rate": 3.1667818936872465e-05,
+      "loss": 0.2641,
       "step": 54
     },
     {
       "epoch": 1.75,
-      "learning_rate": 3.634976249348867e-05,
-      "loss": 0.2519,
+      "learning_rate": 3.0477531003921745e-05,
+      "loss": 0.2588,
       "step": 56
     },
     {
       "epoch": 1.81,
-      "learning_rate": 3.54664934384357e-05,
-      "loss": 0.27,
+      "learning_rate": 2.9274047219007534e-05,
+      "loss": 0.2366,
       "step": 58
     },
     {
       "epoch": 1.88,
-      "learning_rate": 3.456708580912725e-05,
-      "loss": 0.2511,
+      "learning_rate": 2.8060266879980408e-05,
+      "loss": 0.2581,
       "step": 60
     },
     {
       "epoch": 1.94,
-      "learning_rate": 3.365292642693732e-05,
-      "loss": 0.2409,
+      "learning_rate": 2.683911408999169e-05,
+      "loss": 0.2658,
       "step": 62
     },
     {
       "epoch": 2.0,
-      "learning_rate": 3.272542485937369e-05,
-      "loss": 0.2486,
+      "learning_rate": 2.561353071307281e-05,
+      "loss": 0.2474,
       "step": 64
     },
     {
       "epoch": 2.06,
-      "learning_rate": 3.178601124662686e-05,
-      "loss": 0.1282,
+      "learning_rate": 2.4386469286927196e-05,
+      "loss": 0.1347,
       "step": 66
     },
     {
       "epoch": 2.12,
-      "learning_rate": 3.083613409639764e-05,
-      "loss": 0.1339,
+      "learning_rate": 2.3160885910008318e-05,
+      "loss": 0.135,
       "step": 68
     },
     {
       "epoch": 2.19,
-      "learning_rate": 2.9877258050403212e-05,
-      "loss": 0.1175,
+      "learning_rate": 2.1939733120019598e-05,
+      "loss": 0.1242,
       "step": 70
     },
     {
       "epoch": 2.25,
-      "learning_rate": 2.8910861626005776e-05,
-      "loss": 0.109,
+      "learning_rate": 2.0725952780992468e-05,
+      "loss": 0.1286,
       "step": 72
     },
     {
       "epoch": 2.31,
-      "learning_rate": 2.7938434936445945e-05,
-      "loss": 0.12,
+      "learning_rate": 1.9522468996078258e-05,
+      "loss": 0.1252,
       "step": 74
-    },
-    {
-      "epoch": 2.38,
-      "learning_rate": 2.6961477393196126e-05,
-      "loss": 0.1263,
-      "step": 76
-    },
-    {
-      "epoch": 2.44,
-      "learning_rate": 2.598149539397672e-05,
-      "loss": 0.1206,
-      "step": 78
-    },
-    {
-      "epoch": 2.5,
-      "learning_rate": 2.5e-05,
-      "loss": 0.1147,
-      "step": 80
-    },
-    {
-      "epoch": 2.56,
-      "learning_rate": 2.4018504606023293e-05,
-      "loss": 0.1148,
-      "step": 82
-    },
-    {
-      "epoch": 2.62,
-      "learning_rate": 2.303852260680388e-05,
-      "loss": 0.1162,
-      "step": 84
-    },
-    {
-      "epoch": 2.69,
-      "learning_rate": 2.2061565063554064e-05,
-      "loss": 0.1083,
-      "step": 86
-    },
-    {
-      "epoch": 2.75,
-      "learning_rate": 2.1089138373994223e-05,
-      "loss": 0.117,
-      "step": 88
-    },
-    {
-      "epoch": 2.81,
-      "learning_rate": 2.0122741949596797e-05,
-      "loss": 0.0933,
-      "step": 90
-    },
-    {
-      "epoch": 2.88,
-      "learning_rate": 1.9163865903602374e-05,
-      "loss": 0.1052,
-      "step": 92
-    },
-    {
-      "epoch": 2.94,
-      "learning_rate": 1.8213988753373146e-05,
-      "loss": 0.1112,
-      "step": 94
-    },
-    {
-      "epoch": 3.0,
-      "learning_rate": 1.7274575140626318e-05,
-      "loss": 0.1048,
-      "step": 96
-    },
-    {
-      "epoch": 3.06,
-      "learning_rate": 1.6347073573062672e-05,
-      "loss": 0.0463,
-      "step": 98
-    },
-    {
-      "epoch": 3.12,
-      "learning_rate": 1.5432914190872757e-05,
-      "loss": 0.0518,
-      "step": 100
     }
   ],
   "logging_steps": 2,
-  "max_steps": 160,
+  "max_steps": 128,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
-  "save_steps": 20,
-  "total_flos": 1.0844211055626813e+18,
+  "num_train_epochs": 4,
+  "save_steps": 25,
+  "total_flos": 8.20785473084457e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:70b3c37f005a0c19e72439d1eb4b0e077d23fb67b5b2041a4b2911b018eafd1d
+oid sha256:7d54f2c2e57d38a1e8d2652900b46927ec020d8523d9bb88688de54db84bd708
 size 6136