Upload folder using huggingface_hub

README.md

@@ -16,7 +16,7 @@ language: ['pt', 'en', 'es']
 This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2] on the dataset Weni/zeroshot-3.2.0 with the SFT trainer. It is part of the zeroshot project for [Weni](https://weni.ai/).
 
 It achieves the following results on the evaluation set:
-{'eval_loss': 0.3796847462654114, 'eval_runtime': 94.1886, 'eval_samples_per_second': 30.439, 'eval_steps_per_second': 0.956, 'epoch': 0.99}
+{'eval_loss': 0.3921997845172882, 'eval_runtime': 180.2964, 'eval_samples_per_second': 15.902, 'eval_steps_per_second': 0.998, 'epoch': 0.5}
 
 ## Intended uses & limitations
 

checkpoint-100/adapter_config.json

@@ -20,9 +20,9 @@
   "revision": null,
   "target_modules": [
     "q_proj",
+    "k_proj",
     "v_proj",
-    "o_proj",
-    "k_proj"
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_rslora": false

checkpoint-100/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8307659fd0862cf0805b8206dbc2159ba622b3301d90d9634a43e58ab49be69d
+oid sha256:f4a9f21cd881464354a5d422dcae331c3c4db388c18576c0ecb9a93d6bce5fb5
 size 436242776

checkpoint-100/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f7ba7a8726d78998756dd55f553ecc0a377148f2698744401894deb794dff765
+oid sha256:3cdc22fb86973bff2cc05ff2b6c9d57e33f82eaed8c614ce25ca8e457bf79192
 size 872568314

checkpoint-100/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23673dcfe8f26201d1b64968cfb8e7aeafa0dd44591f91dcef9addc46b95c804
-size 15024
+oid sha256:94c603473dc4b52ea920ce532750d7fb82f8a6b1014d2d3bfeecb430ff7916e7
+size 14512

checkpoint-100/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6469a17730051cc187688ed7abb5fd4c8dde61701d8ada08c6a999a9463ac217
-size 15024
+oid sha256:2fb0b5af4969d319c137ef120d1a81b3c1905b6f9244439e660950fd2eabdcd0
+size 14512

checkpoint-100/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad10f9bdd16d4aa7ed3e572ac6474b4271d2e92b701684b91f54abd90b8638ef
+oid sha256:3d2ae6e01a1808af3d88a28dc34a871dccf737b3b52505dba8e2c66818fbd018
 size 1000

checkpoint-100/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.3796847462654114,
+  "best_metric": 0.3921997845172882,
   "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-100",
-  "epoch": 0.9900990099009901,
+  "epoch": 0.49566294919454773,
   "eval_steps": 10,
   "global_step": 100,
   "is_hyper_param_search": false,
@@ -9,118 +9,118 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.1,
-      "eval_loss": 0.6981944441795349,
-      "eval_runtime": 94.6462,
-      "eval_samples_per_second": 30.292,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.05,
+      "eval_loss": 0.6367942690849304,
+      "eval_runtime": 180.8957,
+      "eval_samples_per_second": 15.849,
+      "eval_steps_per_second": 0.995,
       "step": 10
     },
     {
-      "epoch": 0.2,
-      "grad_norm": 0.2654201090335846,
-      "learning_rate": 0.0001961261695938319,
-      "loss": 1.0016,
+      "epoch": 0.1,
+      "grad_norm": 0.34346863627433777,
+      "learning_rate": 0.00019510565162951537,
+      "loss": 0.9774,
       "step": 20
     },
     {
-      "epoch": 0.2,
-      "eval_loss": 0.467332661151886,
-      "eval_runtime": 94.5373,
-      "eval_samples_per_second": 30.327,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.1,
+      "eval_loss": 0.48089343309402466,
+      "eval_runtime": 180.8882,
+      "eval_samples_per_second": 15.85,
+      "eval_steps_per_second": 0.995,
       "step": 20
     },
     {
-      "epoch": 0.3,
-      "eval_loss": 0.434685617685318,
-      "eval_runtime": 94.5899,
-      "eval_samples_per_second": 30.31,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.15,
+      "eval_loss": 0.44581571221351624,
+      "eval_runtime": 180.8305,
+      "eval_samples_per_second": 15.855,
+      "eval_steps_per_second": 0.995,
       "step": 30
     },
     {
-      "epoch": 0.4,
-      "grad_norm": 1.3763986825942993,
-      "learning_rate": 0.0001559192903470747,
-      "loss": 0.4357,
+      "epoch": 0.2,
+      "grad_norm": 0.22729116678237915,
+      "learning_rate": 0.0001529919264233205,
+      "loss": 0.4459,
       "step": 40
     },
     {
-      "epoch": 0.4,
-      "eval_loss": 0.4082697033882141,
-      "eval_runtime": 94.551,
-      "eval_samples_per_second": 30.322,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.2,
+      "eval_loss": 0.4221758544445038,
+      "eval_runtime": 180.8406,
+      "eval_samples_per_second": 15.854,
+      "eval_steps_per_second": 0.995,
       "step": 40
     },
     {
-      "epoch": 0.5,
-      "eval_loss": 0.40642818808555603,
-      "eval_runtime": 94.5831,
-      "eval_samples_per_second": 30.312,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.25,
+      "eval_loss": 0.40922558307647705,
+      "eval_runtime": 181.1791,
+      "eval_samples_per_second": 15.824,
+      "eval_steps_per_second": 0.993,
       "step": 50
     },
     {
-      "epoch": 0.59,
-      "grad_norm": 0.16300354897975922,
-      "learning_rate": 8.954715367323468e-05,
-      "loss": 0.4012,
+      "epoch": 0.3,
+      "grad_norm": 0.27356910705566406,
+      "learning_rate": 8.608268990399349e-05,
+      "loss": 0.4132,
       "step": 60
     },
     {
-      "epoch": 0.59,
-      "eval_loss": 0.39164847135543823,
-      "eval_runtime": 94.6074,
-      "eval_samples_per_second": 30.304,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.3,
+      "eval_loss": 0.4156472384929657,
+      "eval_runtime": 180.7116,
+      "eval_samples_per_second": 15.865,
+      "eval_steps_per_second": 0.996,
       "step": 60
     },
     {
-      "epoch": 0.69,
-      "eval_loss": 0.3856147527694702,
-      "eval_runtime": 94.5883,
-      "eval_samples_per_second": 30.31,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.35,
+      "eval_loss": 0.39848339557647705,
+      "eval_runtime": 180.7856,
+      "eval_samples_per_second": 15.859,
+      "eval_steps_per_second": 0.996,
       "step": 70
     },
     {
-      "epoch": 0.79,
-      "grad_norm": 0.14875428378582,
-      "learning_rate": 2.8066019966134904e-05,
-      "loss": 0.3846,
+      "epoch": 0.4,
+      "grad_norm": 0.1912572681903839,
+      "learning_rate": 2.5685517452260567e-05,
+      "loss": 0.4004,
       "step": 80
     },
     {
-      "epoch": 0.79,
-      "eval_loss": 0.38140159845352173,
-      "eval_runtime": 94.5138,
-      "eval_samples_per_second": 30.334,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.4,
+      "eval_loss": 0.3947867155075073,
+      "eval_runtime": 180.6659,
+      "eval_samples_per_second": 15.869,
+      "eval_steps_per_second": 0.996,
       "step": 80
     },
     {
-      "epoch": 0.89,
-      "eval_loss": 0.3798506557941437,
-      "eval_runtime": 94.6456,
-      "eval_samples_per_second": 30.292,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.45,
+      "eval_loss": 0.39267557859420776,
+      "eval_runtime": 180.6936,
+      "eval_samples_per_second": 15.867,
+      "eval_steps_per_second": 0.996,
       "step": 90
     },
     {
-      "epoch": 0.99,
-      "grad_norm": 3.9488003253936768,
-      "learning_rate": 2.4359497401758024e-07,
-      "loss": 0.3775,
+      "epoch": 0.5,
+      "grad_norm": 0.1869862824678421,
+      "learning_rate": 6.09172980904238e-08,
+      "loss": 0.3929,
       "step": 100
     },
     {
-      "epoch": 0.99,
-      "eval_loss": 0.3796847462654114,
-      "eval_runtime": 94.5333,
-      "eval_samples_per_second": 30.328,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.5,
+      "eval_loss": 0.3921997845172882,
+      "eval_runtime": 180.1721,
+      "eval_samples_per_second": 15.913,
+      "eval_steps_per_second": 0.999,
       "step": 100
     }
   ],
@@ -129,7 +129,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 10,
-  "total_flos": 9.048313688175411e+17,
+  "total_flos": 4.521353175336223e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

checkpoint-100/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66ceb8ab0424702f162d3dcd459022d693533c009fa75ecbe9af10b7fcf8a54d
+oid sha256:397ed785970a10324cddec1fc7a5b8a987e32e5f4a937dc38def4b1cb481ef1b
 size 5176

checkpoint-70/adapter_config.json

@@ -20,9 +20,9 @@
   "revision": null,
   "target_modules": [
     "q_proj",
+    "k_proj",
     "v_proj",
-    "o_proj",
-    "k_proj"
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_rslora": false

checkpoint-70/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3729ade0578702d582ba4201d2056d525c6b98445fc9f287499aa8cc4a04ad0f
+oid sha256:cfb912af6b1796046b92cd31a140d407273ff753ac7f3269fc3c5f3185cb84e1
 size 436242776

checkpoint-70/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f41106fec66c8da66955717a5c0e57087dbb2ce1e9c2786539e9d14f13dba082
+oid sha256:631ee3323927506f442194f35db41e1ecdfb11bea1ce0e91876dfd94f7df30c1
 size 872568314

checkpoint-70/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f23d434a5adf472af32d0d067bd6c7196c5fada3cf7bbec69afdf226737ae9c
-size 15024
+oid sha256:b7a0a43eb1da4e2bdacd5b9613bc246aef81888090db42765afb6f369c67c4c1
+size 14512

checkpoint-70/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f8cd675fa29ceb7104e041bb8b6f813a24babb71dc2874373344d885230ca229
-size 15024
+oid sha256:90cb909835d393126c788ffdccbb0e802fe522bcd647fc16615a348d6c39cfd4
+size 14512

checkpoint-70/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a45b3b845e6140514d22eb89d3d82950a7efd6b6e0c772b6f3e23d8289d822fd
+oid sha256:74915e0aef978c84c56eee8bed06ba18d446da4dd1125eda8fbb469b255811be
 size 1000

checkpoint-70/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.3856147527694702,
+  "best_metric": 0.39848339557647705,
   "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-70",
-  "epoch": 0.693069306930693,
+  "epoch": 0.3469640644361834,
   "eval_steps": 10,
   "global_step": 70,
   "is_hyper_param_search": false,
@@ -9,80 +9,80 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.1,
-      "eval_loss": 0.6981944441795349,
-      "eval_runtime": 94.6462,
-      "eval_samples_per_second": 30.292,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.05,
+      "eval_loss": 0.6367942690849304,
+      "eval_runtime": 180.8957,
+      "eval_samples_per_second": 15.849,
+      "eval_steps_per_second": 0.995,
       "step": 10
     },
     {
-      "epoch": 0.2,
-      "grad_norm": 0.2654201090335846,
-      "learning_rate": 0.0001961261695938319,
-      "loss": 1.0016,
+      "epoch": 0.1,
+      "grad_norm": 0.34346863627433777,
+      "learning_rate": 0.00019510565162951537,
+      "loss": 0.9774,
       "step": 20
     },
     {
-      "epoch": 0.2,
-      "eval_loss": 0.467332661151886,
-      "eval_runtime": 94.5373,
-      "eval_samples_per_second": 30.327,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.1,
+      "eval_loss": 0.48089343309402466,
+      "eval_runtime": 180.8882,
+      "eval_samples_per_second": 15.85,
+      "eval_steps_per_second": 0.995,
       "step": 20
     },
     {
-      "epoch": 0.3,
-      "eval_loss": 0.434685617685318,
-      "eval_runtime": 94.5899,
-      "eval_samples_per_second": 30.31,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.15,
+      "eval_loss": 0.44581571221351624,
+      "eval_runtime": 180.8305,
+      "eval_samples_per_second": 15.855,
+      "eval_steps_per_second": 0.995,
       "step": 30
     },
     {
-      "epoch": 0.4,
-      "grad_norm": 1.3763986825942993,
-      "learning_rate": 0.0001559192903470747,
-      "loss": 0.4357,
+      "epoch": 0.2,
+      "grad_norm": 0.22729116678237915,
+      "learning_rate": 0.0001529919264233205,
+      "loss": 0.4459,
       "step": 40
     },
     {
-      "epoch": 0.4,
-      "eval_loss": 0.4082697033882141,
-      "eval_runtime": 94.551,
-      "eval_samples_per_second": 30.322,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.2,
+      "eval_loss": 0.4221758544445038,
+      "eval_runtime": 180.8406,
+      "eval_samples_per_second": 15.854,
+      "eval_steps_per_second": 0.995,
       "step": 40
     },
     {
-      "epoch": 0.5,
-      "eval_loss": 0.40642818808555603,
-      "eval_runtime": 94.5831,
-      "eval_samples_per_second": 30.312,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.25,
+      "eval_loss": 0.40922558307647705,
+      "eval_runtime": 181.1791,
+      "eval_samples_per_second": 15.824,
+      "eval_steps_per_second": 0.993,
       "step": 50
     },
     {
-      "epoch": 0.59,
-      "grad_norm": 0.16300354897975922,
-      "learning_rate": 8.954715367323468e-05,
-      "loss": 0.4012,
+      "epoch": 0.3,
+      "grad_norm": 0.27356910705566406,
+      "learning_rate": 8.608268990399349e-05,
+      "loss": 0.4132,
       "step": 60
     },
     {
-      "epoch": 0.59,
-      "eval_loss": 0.39164847135543823,
-      "eval_runtime": 94.6074,
-      "eval_samples_per_second": 30.304,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.3,
+      "eval_loss": 0.4156472384929657,
+      "eval_runtime": 180.7116,
+      "eval_samples_per_second": 15.865,
+      "eval_steps_per_second": 0.996,
       "step": 60
     },
     {
-      "epoch": 0.69,
-      "eval_loss": 0.3856147527694702,
-      "eval_runtime": 94.5883,
-      "eval_samples_per_second": 30.31,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.35,
+      "eval_loss": 0.39848339557647705,
+      "eval_runtime": 180.7856,
+      "eval_samples_per_second": 15.859,
+      "eval_steps_per_second": 0.996,
       "step": 70
     }
   ],
@@ -91,7 +91,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 10,
-  "total_flos": 6.30628785351295e+17,
+  "total_flos": 3.162393138828411e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

checkpoint-70/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66ceb8ab0424702f162d3dcd459022d693533c009fa75ecbe9af10b7fcf8a54d
+oid sha256:397ed785970a10324cddec1fc7a5b8a987e32e5f4a937dc38def4b1cb481ef1b
 size 5176

checkpoint-80/adapter_config.json

@@ -20,9 +20,9 @@
   "revision": null,
   "target_modules": [
     "q_proj",
+    "k_proj",
     "v_proj",
-    "o_proj",
-    "k_proj"
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_rslora": false

checkpoint-80/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24e21a859be9d207f149ffc0b46dfcc320ea1ee34732d41bf90a24c1b253b907
+oid sha256:b527b969a7485a4a5babdabe6555b5c383c05ebc9f26e37428a33b13c8b5da9d
 size 436242776

checkpoint-80/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f70660925ff51a9f94e304f49828ef04a284f17d3a7bc460079ba9ce1a8e117e
+oid sha256:b1eec7cb54f027bab8ca6c3381b6ffff8c752180f8c5c51218b412ed94f8b847
 size 872568314

checkpoint-80/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93f8281a861fa369d43b6beb0ae6c67fde2ae747fd1b83daa3e6a92279b33244
-size 15024
+oid sha256:0a801669d16f26e581b00baf98c1e7c4a03b79841796ecdf09044e710d489912
+size 14512

checkpoint-80/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:090bdbfa5ddd772617ed624bd84911da0c7bf8de602f4e422d57d8f9b2a59994
-size 15024
+oid sha256:832aea46110ff9ec827428f726a2265f68e7159fe5184ac4697ecf55087199d1
+size 14512

checkpoint-80/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0af4379a58cfae44f0dbe3cf08fcd1bc98068c05bfa6733ba98cd23f1ebaae09
+oid sha256:2b71e69a6ae8462bc71f9d5c4bc5cfca3ea5f220e53bbe18bfaf849b8d7997dd
 size 1000

checkpoint-80/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.38140159845352173,
+  "best_metric": 0.3947867155075073,
   "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-80",
-  "epoch": 0.7920792079207921,
+  "epoch": 0.3965303593556382,
   "eval_steps": 10,
   "global_step": 80,
   "is_hyper_param_search": false,
@@ -9,95 +9,95 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.1,
-      "eval_loss": 0.6981944441795349,
-      "eval_runtime": 94.6462,
-      "eval_samples_per_second": 30.292,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.05,
+      "eval_loss": 0.6367942690849304,
+      "eval_runtime": 180.8957,
+      "eval_samples_per_second": 15.849,
+      "eval_steps_per_second": 0.995,
       "step": 10
     },
     {
-      "epoch": 0.2,
-      "grad_norm": 0.2654201090335846,
-      "learning_rate": 0.0001961261695938319,
-      "loss": 1.0016,
+      "epoch": 0.1,
+      "grad_norm": 0.34346863627433777,
+      "learning_rate": 0.00019510565162951537,
+      "loss": 0.9774,
       "step": 20
     },
     {
-      "epoch": 0.2,
-      "eval_loss": 0.467332661151886,
-      "eval_runtime": 94.5373,
-      "eval_samples_per_second": 30.327,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.1,
+      "eval_loss": 0.48089343309402466,
+      "eval_runtime": 180.8882,
+      "eval_samples_per_second": 15.85,
+      "eval_steps_per_second": 0.995,
       "step": 20
     },
     {
-      "epoch": 0.3,
-      "eval_loss": 0.434685617685318,
-      "eval_runtime": 94.5899,
-      "eval_samples_per_second": 30.31,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.15,
+      "eval_loss": 0.44581571221351624,
+      "eval_runtime": 180.8305,
+      "eval_samples_per_second": 15.855,
+      "eval_steps_per_second": 0.995,
       "step": 30
     },
     {
-      "epoch": 0.4,
-      "grad_norm": 1.3763986825942993,
-      "learning_rate": 0.0001559192903470747,
-      "loss": 0.4357,
+      "epoch": 0.2,
+      "grad_norm": 0.22729116678237915,
+      "learning_rate": 0.0001529919264233205,
+      "loss": 0.4459,
       "step": 40
     },
     {
-      "epoch": 0.4,
-      "eval_loss": 0.4082697033882141,
-      "eval_runtime": 94.551,
-      "eval_samples_per_second": 30.322,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.2,
+      "eval_loss": 0.4221758544445038,
+      "eval_runtime": 180.8406,
+      "eval_samples_per_second": 15.854,
+      "eval_steps_per_second": 0.995,
       "step": 40
     },
     {
-      "epoch": 0.5,
-      "eval_loss": 0.40642818808555603,
-      "eval_runtime": 94.5831,
-      "eval_samples_per_second": 30.312,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.25,
+      "eval_loss": 0.40922558307647705,
+      "eval_runtime": 181.1791,
+      "eval_samples_per_second": 15.824,
+      "eval_steps_per_second": 0.993,
       "step": 50
     },
     {
-      "epoch": 0.59,
-      "grad_norm": 0.16300354897975922,
-      "learning_rate": 8.954715367323468e-05,
-      "loss": 0.4012,
+      "epoch": 0.3,
+      "grad_norm": 0.27356910705566406,
+      "learning_rate": 8.608268990399349e-05,
+      "loss": 0.4132,
       "step": 60
     },
     {
-      "epoch": 0.59,
-      "eval_loss": 0.39164847135543823,
-      "eval_runtime": 94.6074,
-      "eval_samples_per_second": 30.304,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.3,
+      "eval_loss": 0.4156472384929657,
+      "eval_runtime": 180.7116,
+      "eval_samples_per_second": 15.865,
+      "eval_steps_per_second": 0.996,
       "step": 60
     },
     {
-      "epoch": 0.69,
-      "eval_loss": 0.3856147527694702,
-      "eval_runtime": 94.5883,
-      "eval_samples_per_second": 30.31,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.35,
+      "eval_loss": 0.39848339557647705,
+      "eval_runtime": 180.7856,
+      "eval_samples_per_second": 15.859,
+      "eval_steps_per_second": 0.996,
       "step": 70
     },
     {
-      "epoch": 0.79,
-      "grad_norm": 0.14875428378582,
-      "learning_rate": 2.8066019966134904e-05,
-      "loss": 0.3846,
+      "epoch": 0.4,
+      "grad_norm": 0.1912572681903839,
+      "learning_rate": 2.5685517452260567e-05,
+      "loss": 0.4004,
       "step": 80
     },
     {
-      "epoch": 0.79,
-      "eval_loss": 0.38140159845352173,
-      "eval_runtime": 94.5138,
-      "eval_samples_per_second": 30.334,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.4,
+      "eval_loss": 0.3947867155075073,
+      "eval_runtime": 180.6659,
+      "eval_samples_per_second": 15.869,
+      "eval_steps_per_second": 0.996,
       "step": 80
     }
   ],
@@ -106,7 +106,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 10,
-  "total_flos": 7.216782437357978e+17,
+  "total_flos": 3.617727103190958e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

checkpoint-80/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66ceb8ab0424702f162d3dcd459022d693533c009fa75ecbe9af10b7fcf8a54d
+oid sha256:397ed785970a10324cddec1fc7a5b8a987e32e5f4a937dc38def4b1cb481ef1b
 size 5176

checkpoint-90/adapter_config.json

@@ -20,9 +20,9 @@
   "revision": null,
   "target_modules": [
     "q_proj",
+    "k_proj",
     "v_proj",
-    "o_proj",
-    "k_proj"
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_rslora": false

checkpoint-90/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f0c30fc1ce7eac8619bae4324c083d54fc3ab754951ccedd52fac47403a20743
+oid sha256:656b072afb9540faadac0b0741a7a1048ad82d161f456b99cbe9cb27de8904b5
 size 436242776

checkpoint-90/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:486aa4e54416a204b7db0aeed6777c24c2ff3852690250be9c7cbf8a71f41c44
+oid sha256:b434dc7b4c8b42e026285b4cfd5bba70c7c605b762fc1f40a2497755ac2cc407
 size 872568314

checkpoint-90/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8a254ba0cdafb651978de9121c20a71b149b425dfe94a899e5728c492162cdab
-size 15024
+oid sha256:d1ca7c38f60ccd88483b3a63102bc43b67e4b014bcfc0d88a6a512edcbc12712
+size 14512

checkpoint-90/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d1a2d1d643c84a57098344dddf622cb085a1400d5ef0b7fc0334642388ea389
-size 15024
+oid sha256:2d92ee0aa8cf2e1100b47e0467d2fb7b0bafc09e2325a5eebad03c7f3d262f88
+size 14512

checkpoint-90/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:05a9a94c03ab13518f6155b6d7ecc8a918e6d6e0a99e88f4f661c51a3bc91db3
+oid sha256:0658f8849eb58961b159a204d5c11e639b04f4416306a9ce692d0fa29467f5f4
 size 1000

checkpoint-90/trainer_state.json

@@ -1,7 +1,7 @@
 {
-  "best_metric": 0.3798506557941437,
+  "best_metric": 0.39267557859420776,
   "best_model_checkpoint": "./mistral/01-03-24-Weni-ZeroShot-3.3.18-Mistral-7b-Multilanguage-3.2.0_Zeroshot-2_max_steps-100_batch_16_2024-03-01_ppid_7/checkpoint-90",
-  "epoch": 0.8910891089108911,
+  "epoch": 0.44609665427509293,
   "eval_steps": 10,
   "global_step": 90,
   "is_hyper_param_search": false,
@@ -9,103 +9,103 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.1,
-      "eval_loss": 0.6981944441795349,
-      "eval_runtime": 94.6462,
-      "eval_samples_per_second": 30.292,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.05,
+      "eval_loss": 0.6367942690849304,
+      "eval_runtime": 180.8957,
+      "eval_samples_per_second": 15.849,
+      "eval_steps_per_second": 0.995,
       "step": 10
     },
     {
-      "epoch": 0.2,
-      "grad_norm": 0.2654201090335846,
-      "learning_rate": 0.0001961261695938319,
-      "loss": 1.0016,
+      "epoch": 0.1,
+      "grad_norm": 0.34346863627433777,
+      "learning_rate": 0.00019510565162951537,
+      "loss": 0.9774,
       "step": 20
     },
     {
-      "epoch": 0.2,
-      "eval_loss": 0.467332661151886,
-      "eval_runtime": 94.5373,
-      "eval_samples_per_second": 30.327,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.1,
+      "eval_loss": 0.48089343309402466,
+      "eval_runtime": 180.8882,
+      "eval_samples_per_second": 15.85,
+      "eval_steps_per_second": 0.995,
       "step": 20
     },
     {
-      "epoch": 0.3,
-      "eval_loss": 0.434685617685318,
-      "eval_runtime": 94.5899,
-      "eval_samples_per_second": 30.31,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.15,
+      "eval_loss": 0.44581571221351624,
+      "eval_runtime": 180.8305,
+      "eval_samples_per_second": 15.855,
+      "eval_steps_per_second": 0.995,
       "step": 30
     },
     {
-      "epoch": 0.4,
-      "grad_norm": 1.3763986825942993,
-      "learning_rate": 0.0001559192903470747,
-      "loss": 0.4357,
+      "epoch": 0.2,
+      "grad_norm": 0.22729116678237915,
+      "learning_rate": 0.0001529919264233205,
+      "loss": 0.4459,
       "step": 40
     },
     {
-      "epoch": 0.4,
-      "eval_loss": 0.4082697033882141,
-      "eval_runtime": 94.551,
-      "eval_samples_per_second": 30.322,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.2,
+      "eval_loss": 0.4221758544445038,
+      "eval_runtime": 180.8406,
+      "eval_samples_per_second": 15.854,
+      "eval_steps_per_second": 0.995,
       "step": 40
     },
     {
-      "epoch": 0.5,
-      "eval_loss": 0.40642818808555603,
-      "eval_runtime": 94.5831,
-      "eval_samples_per_second": 30.312,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.25,
+      "eval_loss": 0.40922558307647705,
+      "eval_runtime": 181.1791,
+      "eval_samples_per_second": 15.824,
+      "eval_steps_per_second": 0.993,
       "step": 50
     },
     {
-      "epoch": 0.59,
-      "grad_norm": 0.16300354897975922,
-      "learning_rate": 8.954715367323468e-05,
-      "loss": 0.4012,
+      "epoch": 0.3,
+      "grad_norm": 0.27356910705566406,
+      "learning_rate": 8.608268990399349e-05,
+      "loss": 0.4132,
       "step": 60
     },
     {
-      "epoch": 0.59,
-      "eval_loss": 0.39164847135543823,
-      "eval_runtime": 94.6074,
-      "eval_samples_per_second": 30.304,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.3,
+      "eval_loss": 0.4156472384929657,
+      "eval_runtime": 180.7116,
+      "eval_samples_per_second": 15.865,
+      "eval_steps_per_second": 0.996,
       "step": 60
     },
     {
-      "epoch": 0.69,
-      "eval_loss": 0.3856147527694702,
-      "eval_runtime": 94.5883,
-      "eval_samples_per_second": 30.31,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.35,
+      "eval_loss": 0.39848339557647705,
+      "eval_runtime": 180.7856,
+      "eval_samples_per_second": 15.859,
+      "eval_steps_per_second": 0.996,
       "step": 70
     },
     {
-      "epoch": 0.79,
-      "grad_norm": 0.14875428378582,
-      "learning_rate": 2.8066019966134904e-05,
-      "loss": 0.3846,
+      "epoch": 0.4,
+      "grad_norm": 0.1912572681903839,
+      "learning_rate": 2.5685517452260567e-05,
+      "loss": 0.4004,
       "step": 80
     },
     {
-      "epoch": 0.79,
-      "eval_loss": 0.38140159845352173,
-      "eval_runtime": 94.5138,
-      "eval_samples_per_second": 30.334,
-      "eval_steps_per_second": 0.952,
+      "epoch": 0.4,
+      "eval_loss": 0.3947867155075073,
+      "eval_runtime": 180.6659,
+      "eval_samples_per_second": 15.869,
+      "eval_steps_per_second": 0.996,
       "step": 80
     },
     {
-      "epoch": 0.89,
-      "eval_loss": 0.3798506557941437,
-      "eval_runtime": 94.6456,
-      "eval_samples_per_second": 30.292,
-      "eval_steps_per_second": 0.951,
+      "epoch": 0.45,
+      "eval_loss": 0.39267557859420776,
+      "eval_runtime": 180.6936,
+      "eval_samples_per_second": 15.867,
+      "eval_steps_per_second": 0.996,
       "step": 90
     }
   ],
@@ -114,7 +114,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
   "save_steps": 10,
-  "total_flos": 8.129633583858975e+17,
+  "total_flos": 4.080622701075497e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

checkpoint-90/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66ceb8ab0424702f162d3dcd459022d693533c009fa75ecbe9af10b7fcf8a54d
+oid sha256:397ed785970a10324cddec1fc7a5b8a987e32e5f4a937dc38def4b1cb481ef1b
 size 5176