End of training

README.md

@@ -3,9 +3,23 @@ license: apache-2.0
 base_model: google-t5/t5-small
 tags:
 - generated_from_trainer
+datasets:
+- lilferrit/xsum_t5_distillation
+metrics:
+- rouge
 model-index:
 - name: xsum_aligned_smallT5
-  results: []
+  results:
+  - task:
+      name: Summarization
+      type: summarization
+    dataset:
+      name: lilferrit/xsum_t5_distillation
+      type: lilferrit/xsum_t5_distillation
+    metrics:
+    - name: Rouge1
+      type: rouge
+      value: 28.6381
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -13,7 +27,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 # xsum_aligned_smallT5
 
-This model is a fine-tuned version of [google-t5/t5-small](https://huggingface.co/google-t5/t5-small) on an unknown dataset.
+This model is a fine-tuned version of [google-t5/t5-small](https://huggingface.co/google-t5/t5-small) on the lilferrit/xsum_t5_distillation dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.5258
+- Rouge1: 28.6381
+- Rouge2: 7.1512
+- Rougel: 21.3477
+- Rougelsum: 21.2928
+- Gen Len: 27.92
 
 ## Model description
 

all_results.json

@@ -0,0 +1,18 @@
+{
+    "epoch": 30.77,
+    "eval_gen_len": 27.92,
+    "eval_loss": 2.5258138179779053,
+    "eval_rouge1": 28.6381,
+    "eval_rouge2": 7.1512,
+    "eval_rougeL": 21.3477,
+    "eval_rougeLsum": 21.2928,
+    "eval_runtime": 34.7229,
+    "eval_samples": 100,
+    "eval_samples_per_second": 2.88,
+    "eval_steps_per_second": 2.88,
+    "train_loss": 1.359188232421875,
+    "train_runtime": 123.5443,
+    "train_samples": 100,
+    "train_samples_per_second": 25.902,
+    "train_steps_per_second": 1.619
+}

eval_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 30.77,
+    "eval_gen_len": 27.92,
+    "eval_loss": 2.5258138179779053,
+    "eval_rouge1": 28.6381,
+    "eval_rouge2": 7.1512,
+    "eval_rougeL": 21.3477,
+    "eval_rougeLsum": 21.2928,
+    "eval_runtime": 34.7229,
+    "eval_samples": 100,
+    "eval_samples_per_second": 2.88,
+    "eval_steps_per_second": 2.88
+}

generation_config.json

@@ -1,5 +1,4 @@
 {
-  "_from_model_config": true,
   "decoder_start_token_id": 0,
   "eos_token_id": 1,
   "pad_token_id": 0,

runs/Apr22_22-06-47_nlpg03.cs.washington.edu/events.out.tfevents.1713848986.nlpg03.cs.washington.edu.1675484.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fa79e977bdab6d9494f6c14768772155195d8389fdd4c282fb24510d6161e8b
+size 613

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 30.77,
+    "train_loss": 1.359188232421875,
+    "train_runtime": 123.5443,
+    "train_samples": 100,
+    "train_samples_per_second": 25.902,
+    "train_steps_per_second": 1.619
+}

trainer_state.json

@@ -0,0 +1,44 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 30.76923076923077,
+  "eval_steps": 500,
+  "global_step": 200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 15.38,
+      "grad_norm": 2.1158251762390137,
+      "learning_rate": 0.0001,
+      "loss": 1.8108,
+      "step": 100
+    },
+    {
+      "epoch": 30.77,
+      "grad_norm": 2.3235597610473633,
+      "learning_rate": 0.0,
+      "loss": 0.9076,
+      "step": 200
+    },
+    {
+      "epoch": 30.77,
+      "step": 200,
+      "total_flos": 752998431719424.0,
+      "train_loss": 1.359188232421875,
+      "train_runtime": 123.5443,
+      "train_samples_per_second": 25.902,
+      "train_steps_per_second": 1.619
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 200,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 34,
+  "save_steps": 500,
+  "total_flos": 752998431719424.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}