Model save

README.md

@@ -1,21 +1,22 @@
 ---
-license: other
 base_model: hon9kon9ize/yi-1.5-6b-yub-vocab-expanded
 tags:
-- llama-factory
-- full
+- trl
+- sft
 - generated_from_trainer
+datasets:
+- generator
 model-index:
-- name: cantonesellm-6b
+- name: cantonesellm-cpt-202405
   results: []
 ---
 
 <!-- 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. -->
 
-# cantonesellm-6b
+# cantonesellm-cpt-202405
 
-This model is a fine-tuned version of [hon9kon9ize/yi-1.5-6b-yub-vocab-expanded](https://huggingface.co/hon9kon9ize/yi-1.5-6b-yub-vocab-expanded) on the yue_corpus dataset.
+This model is a fine-tuned version of [hon9kon9ize/yi-1.5-6b-yub-vocab-expanded](https://huggingface.co/hon9kon9ize/yi-1.5-6b-yub-vocab-expanded) on the generator dataset.
 
 ## Model description
 
@@ -35,15 +36,15 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 5e-05
-- train_batch_size: 4
+- train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 32
+- gradient_accumulation_steps: 16
 - total_train_batch_size: 128
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
-- lr_scheduler_warmup_ratio: 0.03
-- num_epochs: 1.0
+- lr_scheduler_warmup_ratio: 0.05
+- num_epochs: 1
 
 ### Training results
 
@@ -52,6 +53,6 @@ The following hyperparameters were used during training:
 ### Framework versions
 
 - Transformers 4.40.2
-- Pytorch 2.2.0+cu121
+- Pytorch 2.1.0+cu118
 - Datasets 2.19.1
 - Tokenizers 0.19.1

all_results.json

@@ -1,8 +1,9 @@
 {
-    "epoch": 0.9994304157964686,
-    "total_flos": 1.2030734967018357e+19,
-    "train_loss": 0.9496028307361081,
-    "train_runtime": 40588.5091,
-    "train_samples_per_second": 2.076,
-    "train_steps_per_second": 0.016
+    "epoch": 0.9991804622193083,
+    "total_flos": 1.0450600815746875e+19,
+    "train_loss": 2.588996330897013,
+    "train_runtime": 133512.32,
+    "train_samples": 384224,
+    "train_samples_per_second": 1.097,
+    "train_steps_per_second": 0.009
 }

model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c317ae80702f365ce4635a377d3c370f61eaa255aa3c2831fbe4b2e69aea5f43
-size 4959450112
+oid sha256:03dc0f2fe5a777636711eb1a111d6d56d4121ea63e4acfccf5a03d91562bc922
+size 4961022976

model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9aefbf25642d48b4cd135ee46bc3bdfff583fa4bb9cf614942500ceb0087f9e
+oid sha256:ca746068ea80ba06dc5e5b6b8f4bb696254a8b56f366ec0722bdf573882061cc
 size 4976802816

model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:034334f303ff192f0868f6c726fff67630852297ff5134a9ebd2fc1fefa1e619
-size 2239329256
+oid sha256:d731aa325a7e038e9c8ac14022b4d1b00ab723b739653cf6aaa2ecedc3538f01
+size 2240902120

model.safetensors.index.json

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_size": 12175548416
+    "total_size": 12178694144
   },
   "weight_map": {
     "lm_head.weight": "model-00003-of-00003.safetensors",

train_results.json

@@ -1,8 +1,9 @@
 {
-    "epoch": 0.9994304157964686,
-    "total_flos": 1.2030734967018357e+19,
-    "train_loss": 0.9496028307361081,
-    "train_runtime": 40588.5091,
-    "train_samples_per_second": 2.076,
-    "train_steps_per_second": 0.016
+    "epoch": 0.9991804622193083,
+    "total_flos": 1.0450600815746875e+19,
+    "train_loss": 2.588996330897013,
+    "train_runtime": 133512.32,
+    "train_samples": 384224,
+    "train_samples_per_second": 1.097,
+    "train_steps_per_second": 0.009
 }