Upload 13 files

README.md

@@ -6,7 +6,7 @@ datasets:
 metrics:
 - rouge
 model-index:
-- name: small-3-6-t
+- name: small-3-6
   results:
   - task:
       name: Summarization
@@ -20,22 +20,22 @@ model-index:
     metrics:
     - name: Rouge1
       type: rouge
-      value: 31.7432
+      value: 31.8175
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# small-3-6-t
+# small-3-6
 
-This model is a fine-tuned version of [models/small-3-6](https://huggingface.co/models/small-3-6) on the xsum dataset.
+This model is a fine-tuned version of [x/small-3-6/](https://huggingface.co/x/small-3-6/) on the xsum dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.1322
-- Rouge1: 31.7432
-- Rouge2: 10.077
-- Rougel: 24.8565
-- Rougelsum: 24.8444
-- Gen Len: 28.8587
+- Loss: 2.1335
+- Rouge1: 31.8175
+- Rouge2: 10.1063
+- Rougel: 24.9237
+- Rougelsum: 24.9165
+- Gen Len: 28.4315
 
 ## Model description
 
@@ -55,10 +55,10 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
-- train_batch_size: 32
-- eval_batch_size: 16
+- train_batch_size: 4
+- eval_batch_size: 4
 - seed: 42
-- gradient_accumulation_steps: 2
+- gradient_accumulation_steps: 16
 - total_train_batch_size: 64
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: constant
@@ -71,6 +71,6 @@ The following hyperparameters were used during training:
 ### Framework versions
 
 - Transformers 4.27.0.dev0
-- Pytorch 1.12.1+cu113
-- Datasets 2.10.0
-- Tokenizers 0.13.2
+- Pytorch 1.13.0+cu117
+- Datasets 2.7.1
+- Tokenizers 0.12.1

all_results.json

@@ -1,18 +1,18 @@
 {
     "epoch": 3.0,
-    "eval_gen_len": 28.858656307936787,
-    "eval_loss": 2.132248878479004,
-    "eval_rouge1": 31.7432,
-    "eval_rouge2": 10.077,
-    "eval_rougeL": 24.8565,
-    "eval_rougeLsum": 24.8444,
-    "eval_runtime": 553.66,
+    "eval_gen_len": 28.431535269709542,
+    "eval_loss": 2.1334991455078125,
+    "eval_rouge1": 31.8175,
+    "eval_rouge2": 10.1063,
+    "eval_rougeL": 24.9237,
+    "eval_rougeLsum": 24.9165,
+    "eval_runtime": 1009.3627,
     "eval_samples": 11327,
-    "eval_samples_per_second": 20.458,
-    "eval_steps_per_second": 1.279,
-    "train_loss": 2.37733912099011,
-    "train_runtime": 8397.5704,
+    "eval_samples_per_second": 11.222,
+    "eval_steps_per_second": 2.806,
+    "train_loss": 2.374575984090233,
+    "train_runtime": 14184.3644,
     "train_samples": 204017,
-    "train_samples_per_second": 72.884,
-    "train_steps_per_second": 1.139
+    "train_samples_per_second": 43.15,
+    "train_steps_per_second": 0.674
 }

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "models/small-3-6",
+  "_name_or_path": "x/small-3-6/",
   "architectures": [
     "T5ForConditionalGeneration"
   ],

eval_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 3.0,
-    "eval_gen_len": 28.858656307936787,
-    "eval_loss": 2.132248878479004,
-    "eval_rouge1": 31.7432,
-    "eval_rouge2": 10.077,
-    "eval_rougeL": 24.8565,
-    "eval_rougeLsum": 24.8444,
-    "eval_runtime": 553.66,
+    "eval_gen_len": 28.431535269709542,
+    "eval_loss": 2.1334991455078125,
+    "eval_rouge1": 31.8175,
+    "eval_rouge2": 10.1063,
+    "eval_rougeL": 24.9237,
+    "eval_rougeLsum": 24.9165,
+    "eval_runtime": 1009.3627,
     "eval_samples": 11327,
-    "eval_samples_per_second": 20.458,
-    "eval_steps_per_second": 1.279
+    "eval_samples_per_second": 11.222,
+    "eval_steps_per_second": 2.806
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:682d6e315377557db86cd39f41c08b2f2a3761cecb5e3b5441b3f7afa4bae9ea
-size 270131021
+oid sha256:ca5d0d9b6f61d273138c4dee5e168e949a587e4d43422dd8f236e7d4a3bd181b
+size 270132533

spiece.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
+size 791656

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "train_loss": 2.37733912099011,
-    "train_runtime": 8397.5704,
+    "train_loss": 2.374575984090233,
+    "train_runtime": 14184.3644,
     "train_samples": 204017,
-    "train_samples_per_second": 72.884,
-    "train_steps_per_second": 1.139
+    "train_samples_per_second": 43.15,
+    "train_steps_per_second": 0.674
 }

trainer_state.json

@@ -1,8 +1,8 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 3.0,
-  "global_step": 9564,
+  "epoch": 2.999745123027154,
+  "global_step": 9561,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -10,130 +10,130 @@
     {
       "epoch": 0.16,
       "learning_rate": 0.0001,
-      "loss": 2.5455,
+      "loss": 2.5416,
       "step": 500
     },
     {
       "epoch": 0.31,
       "learning_rate": 0.0001,
-      "loss": 2.4773,
+      "loss": 2.4758,
       "step": 1000
     },
     {
       "epoch": 0.47,
       "learning_rate": 0.0001,
-      "loss": 2.4581,
+      "loss": 2.4544,
       "step": 1500
     },
     {
       "epoch": 0.63,
       "learning_rate": 0.0001,
-      "loss": 2.4289,
+      "loss": 2.4235,
       "step": 2000
     },
     {
       "epoch": 0.78,
       "learning_rate": 0.0001,
-      "loss": 2.4264,
+      "loss": 2.4206,
       "step": 2500
     },
     {
       "epoch": 0.94,
       "learning_rate": 0.0001,
-      "loss": 2.4111,
+      "loss": 2.4084,
       "step": 3000
     },
     {
       "epoch": 1.1,
       "learning_rate": 0.0001,
-      "loss": 2.3918,
+      "loss": 2.3928,
       "step": 3500
     },
     {
-      "epoch": 1.25,
+      "epoch": 1.26,
       "learning_rate": 0.0001,
-      "loss": 2.373,
+      "loss": 2.3703,
       "step": 4000
     },
     {
       "epoch": 1.41,
       "learning_rate": 0.0001,
-      "loss": 2.3684,
+      "loss": 2.3678,
       "step": 4500
     },
     {
       "epoch": 1.57,
       "learning_rate": 0.0001,
-      "loss": 2.364,
+      "loss": 2.3605,
       "step": 5000
     },
     {
       "epoch": 1.73,
       "learning_rate": 0.0001,
-      "loss": 2.3606,
+      "loss": 2.3572,
       "step": 5500
     },
     {
       "epoch": 1.88,
       "learning_rate": 0.0001,
-      "loss": 2.3512,
+      "loss": 2.3473,
       "step": 6000
     },
     {
       "epoch": 2.04,
       "learning_rate": 0.0001,
-      "loss": 2.3383,
+      "loss": 2.3415,
       "step": 6500
     },
     {
       "epoch": 2.2,
       "learning_rate": 0.0001,
-      "loss": 2.3184,
+      "loss": 2.3158,
       "step": 7000
     },
     {
       "epoch": 2.35,
       "learning_rate": 0.0001,
-      "loss": 2.3238,
+      "loss": 2.3191,
       "step": 7500
     },
     {
       "epoch": 2.51,
       "learning_rate": 0.0001,
-      "loss": 2.3079,
+      "loss": 2.3028,
       "step": 8000
     },
     {
       "epoch": 2.67,
       "learning_rate": 0.0001,
-      "loss": 2.3105,
+      "loss": 2.3086,
       "step": 8500
     },
     {
       "epoch": 2.82,
       "learning_rate": 0.0001,
-      "loss": 2.3148,
+      "loss": 2.3124,
       "step": 9000
     },
     {
       "epoch": 2.98,
       "learning_rate": 0.0001,
-      "loss": 2.3072,
+      "loss": 2.3034,
       "step": 9500
     },
     {
       "epoch": 3.0,
-      "step": 9564,
-      "total_flos": 1.916909651434537e+17,
-      "train_loss": 2.37733912099011,
-      "train_runtime": 8397.5704,
-      "train_samples_per_second": 72.884,
-      "train_steps_per_second": 1.139
+      "step": 9561,
+      "total_flos": 1.5154927442907955e+17,
+      "train_loss": 2.374575984090233,
+      "train_runtime": 14184.3644,
+      "train_samples_per_second": 43.15,
+      "train_steps_per_second": 0.674
     }
   ],
-  "max_steps": 9564,
+  "max_steps": 9561,
   "num_train_epochs": 3,
-  "total_flos": 1.916909651434537e+17,
+  "total_flos": 1.5154927442907955e+17,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1b06acb0c42c995c39614ca9c1505acc13befd60ef91232c34d33675a0e2843b
-size 3631
+oid sha256:c8115bce279b297a884a2fc911eea452fbe3fd1f655c5e468581ac9fe6fa7120
+size 3643