Training in progress, step 500

adapter_config.json

@@ -1,6 +1,6 @@
 {
   "auto_mapping": null,
-  "base_model_name_or_path": "vilm/vietcuna-7b-v3",
+  "base_model_name_or_path": "vlsp-2023-vllm/hoa-7b",
   "inference_mode": true,
   "num_attention_heads": 32,
   "num_layers": 30,
@@ -12,5 +12,5 @@
   "revision": null,
   "task_type": "CAUSAL_LM",
   "token_dim": 4096,
-  "tokenizer_name_or_path": "vilm/vietcuna-7b-v3"
+  "tokenizer_name_or_path": "vlsp-2023-vllm/hoa-7b"
 }

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2a34b7f68b08f67e03036273d978c1851859bd4e2ecf3d64c6c92c1ddf00a941
+oid sha256:0be013c72ed6ebf026473764bc51c1e7256ee7ab1a4ebd98dd30078d2af6ee7a
 size 131837

added_tokens.json

@@ -0,0 +1,8 @@
+{
+  "</s>": 2,
+  "<mask>": 4,
+  "<pad>": 1,
+  "<s>": 0,
+  "<unk>": 3,
+  "<|endoftext|>": 5
+}

special_tokens_map.json

@@ -1,6 +1,6 @@
 {
-  "bos_token": "<s>",
-  "eos_token": "</s>",
-  "pad_token": "</s>",
-  "unk_token": "<unk>"
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
 }

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f07e88418b331418961de843cab2dafef7ef384f772a31d90d8442660bd9100
-size 14500736
+oid sha256:c8fc6581f226efaaaeb3170f303fa6c7c6221f450c6eb02799511c6f4767fc3f
+size 2074778

tokenizer_config.json

@@ -2,7 +2,7 @@
   "add_prefix_space": false,
   "added_tokens_decoder": {
     "0": {
-      "content": "<unk>",
+      "content": "<s>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -10,7 +10,7 @@
       "special": true
     },
     "1": {
-      "content": "<s>",
+      "content": "<pad>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -26,7 +26,23 @@
       "special": true
     },
     "3": {
-      "content": "<pad>",
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<|endoftext|>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -35,11 +51,10 @@
     }
   },
   "additional_special_tokens": [],
-  "bos_token": "<s>",
-  "clean_up_tokenization_spaces": false,
-  "eos_token": "</s>",
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
   "model_max_length": 1000000000000000019884624838656,
-  "pad_token": "</s>",
-  "tokenizer_class": "BloomTokenizer",
-  "unk_token": "<unk>"
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ab8534a9b09b30ecb2e2ac9545ee742a45697d1686e72098ac0f7618119a891
-size 4091
+oid sha256:97e9604bfdc6f4450f03c3979a9d3505a163f0c24909cfc529c361cd91aad1e7
+size 4027