Training in progress, step 200

adapter_config.json

@@ -15,7 +15,8 @@
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": [
-    "embed_tokens",
+    "embed_in",
+    "embed_out",
     "lm_head"
   ],
   "peft_type": "LORA",
@@ -23,10 +24,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "query_key_value",
-    "dense_4h_to_h",
+    "dense",
     "dense_h_to_4h",
-    "dense"
+    "query_key_value",
+    "dense_4h_to_h"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e44ce263e6fd885f50d82ca515b9325375b43ee36ededb75acf161ce88bc2e41
-size 48
+oid sha256:9add6957d0638490738073dec6d51b0fc0f74d099bfc1fc6c65dc528d628492f
+size 159266376

config.json

@@ -22,7 +22,7 @@
   "rotary_emb_base": 10000,
   "rotary_pct": 0.25,
   "tie_word_embeddings": false,
-  "torch_dtype": "bfloat16",
+  "torch_dtype": "float16",
   "transformers_version": "4.41.2",
   "use_cache": false,
   "use_parallel_residual": true,

tokenizer.json

@@ -236,6 +236,24 @@
       "rstrip": false,
       "normalized": false,
       "special": true
+    },
+    {
+      "id": 50278,
+      "content": "[INST]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
+    },
+    {
+      "id": 50279,
+      "content": "[/INST]",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": false
     }
   ],
   "normalizer": {

tokenizer_config.json

@@ -210,6 +210,22 @@
       "rstrip": false,
       "single_word": false,
       "special": true
+    },
+    "50278": {
+      "content": "[INST]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50279": {
+      "content": "[/INST]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
     }
   },
   "bos_token": "<|endoftext|>",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8807c1d78d4dc6797f28372f858f1bc8000fbf68ee4085845fa850083e832b42
+oid sha256:0169f40afff071e757b0e090b369d1620ba5cf0efeaea3a91b4e744854f175d1
 size 6072