Upload folder using huggingface_hub (#3)

- Upload folder using huggingface_hub (da2d36ecd42472c355d88cbc9d274b8a2d0f0fa8)
- Update README.md (e99398594b9dd54097e7312db24a907ce9fbcefb)

added_tokens.json

@@ -1,8 +1,9 @@
 {
-  "</CARD>": 49153,
-  "</CARD_SUMMARY>": 49155,
-  "<CARD>": 49152,
-  "<CARD_SUMMARY>": 49154,
-  "<DATASET_CARD>": 49156,
-  "<MODEL_CARD>": 49157
+  "</CARD>": 49154,
+  "</CARD_SUMMARY>": 49156,
+  "<CARD>": 49153,
+  "<CARD_SUMMARY>": 49155,
+  "<DATASET_CARD>": 49157,
+  "<MODEL_CARD>": 49158,
+  "[PAD]": 49152
 }

chat_template.jinja

@@ -0,0 +1 @@
+{% for message in messages %}{% if message['role'] == 'user' %}<CARD>{{ message['content'] }}</CARD>{% elif message['role'] == 'assistant' %}<CARD_SUMMARY>{{ message['content'] }}</CARD_SUMMARY>{% endif %}{% endfor %}{% if not add_generation_prompt %}<|endoftext|>{% endif %}{% if add_generation_prompt %}<CARD_SUMMARY>{% endif %}

config.json

@@ -1,5 +1,4 @@
 {
-  "_name_or_path": "HuggingFaceTB/SmolLM2-360M",
   "architectures": [
     "LlamaForCausalLM"
   ],
@@ -26,7 +25,7 @@
   "rope_theta": 100000,
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.48.3",
+  "transformers_version": "4.52.3",
   "use_cache": true,
-  "vocab_size": 49158
+  "vocab_size": 49159
 }

generation_config.json

@@ -2,5 +2,5 @@
   "_from_model_config": true,
   "bos_token_id": 0,
   "eos_token_id": 0,
-  "transformers_version": "4.48.3"
+  "transformers_version": "4.52.3"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:458b1458e6d5eb6c17779eff4a8f82e6df895ad0ce8fc06a8f04f0dc7a3d4377
-size 723686432
+oid sha256:a5ad1d90198dfce88e6ee1f90539332d12d087f3a59da4f6abdf436813ff622c
+size 723688352

runs/May28_17-48-55_r-davanstrien-jupyterlab-svh3scip-91c05-sx22q/events.out.tfevents.1748447366.r-davanstrien-jupyterlab-svh3scip-91c05-sx22q.1156.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e473849f0b893c6dad386796492a653251310f5d85bb1a862d98be3b90c16e7
+size 18148

special_tokens_map.json

@@ -57,7 +57,13 @@
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": "<|endoftext|>",
+  "pad_token": {
+    "content": "[PAD]",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
   "unk_token": {
     "content": "<|endoftext|>",
     "lstrip": false,

tokenizer.json

@@ -1,11 +1,6 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 8192,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
+  "truncation": null,
   "padding": null,
   "added_tokens": [
     {
@@ -163,7 +158,7 @@
     },
     {
       "id": 49152,
-      "content": "<CARD>",
+      "content": "[PAD]",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -172,7 +167,7 @@
     },
     {
       "id": 49153,
-      "content": "</CARD>",
+      "content": "<CARD>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -181,7 +176,7 @@
     },
     {
       "id": 49154,
-      "content": "<CARD_SUMMARY>",
+      "content": "</CARD>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -190,7 +185,7 @@
     },
     {
       "id": 49155,
-      "content": "</CARD_SUMMARY>",
+      "content": "<CARD_SUMMARY>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -199,7 +194,7 @@
     },
     {
       "id": 49156,
-      "content": "<DATASET_CARD>",
+      "content": "</CARD_SUMMARY>",
       "single_word": false,
       "lstrip": false,
       "rstrip": false,
@@ -208,6 +203,15 @@
     },
     {
       "id": 49157,
+      "content": "<DATASET_CARD>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    {
+      "id": 49158,
       "content": "<MODEL_CARD>",
       "single_word": false,
       "lstrip": false,

tokenizer_config.json

@@ -138,7 +138,7 @@
       "special": true
     },
     "49152": {
-      "content": "<CARD>",
+      "content": "[PAD]",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -146,7 +146,7 @@
       "special": true
     },
     "49153": {
-      "content": "</CARD>",
+      "content": "<CARD>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -154,7 +154,7 @@
       "special": true
     },
     "49154": {
-      "content": "<CARD_SUMMARY>",
+      "content": "</CARD>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -162,7 +162,7 @@
       "special": true
     },
     "49155": {
-      "content": "</CARD_SUMMARY>",
+      "content": "<CARD_SUMMARY>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -170,7 +170,7 @@
       "special": true
     },
     "49156": {
-      "content": "<DATASET_CARD>",
+      "content": "</CARD_SUMMARY>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -178,6 +178,14 @@
       "special": true
     },
     "49157": {
+      "content": "<DATASET_CARD>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "49158": {
       "content": "<MODEL_CARD>",
       "lstrip": false,
       "normalized": false,
@@ -195,12 +203,11 @@
     "<MODEL_CARD>"
   ],
   "bos_token": "<|endoftext|>",
-  "chat_template": "{% for message in messages %}{% if message['role'] == 'user' %}<CARD>{{ message['content'] }}</CARD>{% elif message['role'] == 'assistant' %}<CARD_SUMMARY>{{ message['content'] }}</CARD_SUMMARY>{% endif %}{% endfor %}{% if not add_generation_prompt %}<|endoftext|>{% endif %}{% if add_generation_prompt %}<CARD_SUMMARY>{% endif %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|endoftext|>",
   "extra_special_tokens": {},
   "model_max_length": 8192,
-  "pad_token": "<|endoftext|>",
+  "pad_token": "[PAD]",
   "tokenizer_class": "GPT2Tokenizer",
   "unk_token": "<|endoftext|>",
   "vocab_size": 49152

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc065d067b0b20c9ee90c61037b138a771ed147cc184d5523c1e1e87c5446736
-size 5688
+oid sha256:75bc5c4c8f5f8eed5a885964a0e2fea3ca3737607c6f9890a35ef46b516d96d9
+size 6097