Training in progress, step 100

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/home/ubuntu/hieu.nn/LangProject/alignment-handbook/outputs/models/pythia-1.4b-sft-full/final-model",
+  "_name_or_path": "nnheui/pythia-1.4b-sft-full",
   "architectures": [
     "GPTNeoXForCausalLM"
   ],
@@ -23,7 +23,7 @@
   "rotary_pct": 0.25,
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.38.2",
+  "transformers_version": "4.40.0",
   "use_cache": false,
   "use_parallel_residual": true,
   "vocab_size": 50304

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7985dff3ea63585b23c86c702954f01e9d87d382f21dd6aabd582482c423b6c8
+oid sha256:635f619ab31b0b3348b39a70a64e85d9f0a51fc44cabb2a9a855bcfaf593745d
 size 2829330208

runs/Jul08_01-33-05_42dbe5cf9ed4/events.out.tfevents.1720402951.42dbe5cf9ed4.841582.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a57b5ce5b1aca85d8570a2bd960cfc0ae305300df9cb70050a1670032b5c507c
+size 13908

tokenizer.json

@@ -239,10 +239,30 @@
     "use_regex": true
   },
   "post_processor": {
-    "type": "ByteLevel",
-    "add_prefix_space": false,
-    "trim_offsets": true,
-    "use_regex": true
+    "type": "TemplateProcessing",
+    "single": [
+      {
+        "Sequence": {
+          "id": "A",
+          "type_id": 0
+        }
+      }
+    ],
+    "pair": [
+      {
+        "Sequence": {
+          "id": "A",
+          "type_id": 0
+        }
+      },
+      {
+        "Sequence": {
+          "id": "B",
+          "type_id": 1
+        }
+      }
+    ],
+    "special_tokens": {}
   },
   "decoder": {
     "type": "ByteLevel",
@@ -258,6 +278,7 @@
     "end_of_word_suffix": null,
     "fuse_unk": false,
     "byte_fallback": false,
+    "ignore_merges": false,
     "vocab": {
       "<|endoftext|>": 0,
       "<|padding|>": 1,

tokenizer_config.json

@@ -1,4 +1,6 @@
 {
+  "add_bos_token": false,
+  "add_eos_token": false,
   "add_prefix_space": false,
   "added_tokens_decoder": {
     "0": {

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:40d40e1ead2e5443b55473419d9b5be7870cb8fe5679d6a416b4a119812c2701
-size 6200
+oid sha256:794aae45412f58e207f7e57b16d0915e74c54e21f8feb1bcf71aef160294272c
+size 6264