Training in progress, step 500

adapter_config.json

@@ -1,6 +1,9 @@
 {
   "alpha_pattern": {},
-  "auto_mapping": null,
+  "auto_mapping": {
+    "base_model_class": "LlamaForCausalLM",
+    "parent_library": "transformers.models.llama.modeling_llama"
+  },
   "base_model_name_or_path": "sarvamai/sarvam-2b",
   "bias": "none",
   "eva_config": null,
@@ -23,15 +26,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "down_proj",
     "o_proj",
-    "k_proj",
-    "up_proj",
-    "gate_proj",
     "v_proj",
-    "q_proj"
+    "q_proj",
+    "k_proj",
+    "lm_head"
   ],
-  "task_type": "CAUSAL_LM",
+  "task_type": null,
   "use_dora": false,
   "use_rslora": false
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:848415773217a81110ff0535cf841fe3a6b19fb999a8c79567100d2cc32fa8a2
-size 941420376
+oid sha256:e12a3e1de962f984fe7e1cbae3a22e55daa33f9b9e45d0adacaae8dcc0edb5c1
+size 399673504

special_tokens_map.json

@@ -1,13 +1,4 @@
 {
-  "additional_special_tokens": [
-    {
-      "content": "[PAD]",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false
-    }
-  ],
   "bos_token": {
     "content": "<s>",
     "lstrip": false,

tokenizer_config.json

@@ -32860,18 +32860,14 @@
       "special": true
     }
   },
-  "additional_special_tokens": [
-    "[PAD]"
-  ],
   "bos_token": "<s>",
   "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = false %}{% endif %}\n{% for message in loop_messages %}\n{% if message['role'] not in ['user', 'assistant', 'tool_calls'] %}\n{{ raise_exception('Invalid role: ' + message['role'] + '. Must be user, assistant, or tool_calls.') }}\n{% endif %}\n{% if loop.index0 == 0 and system_message != false %}\n{% set content = '<<SYS>>\n' + system_message + '\n<</SYS>>\n\n' + message['content'] %}\n{% else %}\n{% set content = message['content'] %}\n{% endif %}\n{% if message['role'] == 'user' %}\n{{ bos_token + '[INST] ' + content.strip() + ' [/INST]' }}\n{% elif message['role'] == 'assistant' %}\n{{ ' ' + content.strip() + ' ' + eos_token }}\n{% elif message['role'] == 'tool_calls' %}\n{{ ' [TOOL_CALLS] ' + content.strip() + ' [/TOOL_CALLS] ' }}\n{% endif %}\n{% endfor %}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "</s>",
   "extra_special_tokens": {},
   "legacy": false,
-  "model_max_length": 8192,
+  "model_max_length": 1000000000000000019884624838656,
   "pad_token": "[PAD]",
-  "padding_side": "right",
   "sp_model_kwargs": {},
   "spaces_between_special_tokens": false,
   "tokenizer_class": "LlamaTokenizer",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba593b17db62769f48cad2733568a19f9d173e99ae9d8956ed8b3f4f7f9c8598
-size 5496
+oid sha256:607f46bf18c2b00720eff6ab5f1c2f2f3fa1e25b7abaa8efd8f6682b77a3b582
+size 5240