Post release fixes (#2581)

* fix missing kwarg on child

* make the runpod test shorter

* update docs

* rename runpod test json file

* typing fixes and ordering of doc

.runpod/test-input.json

@@ -12,22 +12,22 @@
       "base_model": "HuggingFaceTB/SmolLM2-135M",
       "model_type": "AutoModelForCausalLM",
       "tokenizer_type": "AutoTokenizer",
-      "load_in_8bit": true,
-      "load_in_4bit": false,
+      "load_in_4bit": true,
       "strict": false,
       "datasets": [
         {
           "path": "mhenrichsen/alpaca_2k_test",
-          "type": "alpaca"
+          "type": "alpaca",
+          "split": "train[:10%]"
         }
       ],
-      "val_set_size": 0.05,
+      "val_set_size": 0.02,
       "output_dir": "./outputs/lora-out",
       "sequence_len": 4096,
       "sample_packing": true,
       "eval_sample_packing": false,
       "pad_to_sequence_len": true,
-      "adapter": "lora",
+      "adapter": "qlora",
       "lora_r": 32,
       "lora_alpha": 64,
       "lora_dropout": 0.05,
@@ -36,8 +36,8 @@
         "embed_tokens",
         "lm_head"
       ],
-      "gradient_accumulation_steps": 4,
-      "micro_batch_size": 2,
+      "gradient_accumulation_steps": 2,
+      "micro_batch_size": 1,
       "num_epochs": 1,
       "optimizer": "adamw_torch_fused",
       "lr_scheduler": "cosine",
@@ -56,7 +56,8 @@
       "weight_decay": 0.0,
       "special_tokens": {
         "pad_token": "<|endoftext|>"
-      }
+      },
+      "max_steps": 20
     },
     "timeout": 100000
   },