Training in progress, step 100

adapter_config.json

@@ -19,23 +19,16 @@
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 64,
+  "r": 8,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "gate_proj",
     "out_proj",
-    "linear_1",
-    "v_proj",
-    "o_proj",
     "k_proj",
+    "v_proj",
     "q_proj",
-    "down_proj",
-    "fc1",
-    "linear_2",
-    "lm_head",
-    "up_proj",
-    "fc2"
+    "fc2",
+    "fc1"
   ],
   "task_type": null,
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d254a2975c64a962d2fdd107197dcf63b2fadfb99f231d1395d8223767d1eddf
-size 1290994752
+oid sha256:6363d9809931b9a7023a620f4cfc1c9e6eaf214ab5f5e205c227875189a99120
+size 491454648

runs/Oct24_20-51-57_dcc93fd67f6b/events.out.tfevents.1729803155.dcc93fd67f6b.2806.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7bedc0b8593b31760d2900be97495b6e1216082c0f64e3300f060e7fde3ec58
+size 20084

runs/Oct24_20-54-07_dcc93fd67f6b/events.out.tfevents.1729803849.dcc93fd67f6b.2806.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a854e76032bec47256dd32cb3a29c190e3e79da8221a6fcd313475d58e08863
+size 10086

runs/Oct24_21-07-27_dcc93fd67f6b/events.out.tfevents.1729804059.dcc93fd67f6b.2806.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68b1a57d80086299da73db5092231d750c6bd984a5d9e0d4decda50bad14a7a3
+size 10086

runs/Oct24_21-13-13_dcc93fd67f6b/events.out.tfevents.1729804394.dcc93fd67f6b.2806.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa96c4657f77b8a5211400e0e469bdb0316091c408be5e6a4994bf04e58a9b1d
+size 10086

runs/Oct24_21-32-25_dcc93fd67f6b/events.out.tfevents.1729805546.dcc93fd67f6b.2806.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a53b8ce2a79815cf02a23856fd8530c8d9d0c4b0af9edfba83adff20ce499231
+size 10086

runs/Oct24_21-39-07_dcc93fd67f6b/events.out.tfevents.1729805948.dcc93fd67f6b.2806.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:316aaf221423463b764f1d7e587975cef65153d339dcaa0ddf0afb2350920958
+size 10086

runs/Oct24_21-44-10_dcc93fd67f6b/events.out.tfevents.1729806251.dcc93fd67f6b.2806.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8009f726076524d9d506baf0fbdf04b1c19c037a7d5697ad52eec823051a13cd
+size 14574

tokenizer_config.json

@@ -45,7 +45,7 @@
     }
   },
   "bos_token": "<s>",
-  "chat_template": "A chat between a curious user and an artificial intelligence assistant.                         The assistant gives helpful, detailed, and polite answers to the user's questions.                         {% for message in messages %}{% if message['role'] == 'user' %}                        USER: {% else %}ASSISTANT: {% endif %}{% for item in message['content'] %}{% if item['type'] == 'text' %}{{ item['text'] }}{% elif item['type'] == 'image' %}<image>{% endif %}{% endfor %}                        {% if message['role'] == 'user' %} {% else %}{{eos_token}}{% endif %}{% endfor %}",
+  "chat_template": "\nA chat between a curious user and an artificial intelligence assistant.\nThe assistant gives helpful, detailed, and polite answers to the user's questions.\n{% for message in messages %}\n{% if message['from'] == 'USER' %}USER: {{ message['value'] }}\n{% else %}ASSISTANT: {{ message['value'] }}{{ eos_token }}\n{% endif %}\n{% endfor %}\n",
   "clean_up_tokenization_spaces": false,
   "eos_token": "</s>",
   "legacy": false,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da24779be1f236da866a138c7aae13fee51b74f9cfa6c88210c57a044876b592
+oid sha256:a73de8b8ef147a6ccb1932736fed68f78b4ebae729f8689b2c77f20b60cf6c05
 size 5560