improve support for customized dataset for bench evals

src/axolotl/utils/callbacks.py

@@ -124,29 +124,21 @@ def bench_eval_callback_factory(trainer, tokenizer):
         tokenizer("G", add_special_tokens=False).input_ids[0],
     ]
     bench_split = "eval"
-    if trainer.args.bench_dataset == "sampled":
 
-        def transform_subject(example):
-            # Split on ':' and trim whitespace
-            parts = example["subject"].split(":")
-            first_part = (
-                parts[0].strip().lower().replace("-", "_")
-            )  # Lowercase the first part
-            second_part = (
-                parts[1].strip().replace("-", "_") if len(parts) > 1 else "all"
-            )  # Replace hyphens with underscores
+    def transform_bench_subject(example):
+        # Split on ':' and trim whitespace
+        parts = example["subject"].split(":")
+        first_part = (
+            parts[0].strip().lower().replace("-", "_")
+        )  # Lowercase the first part
+        second_part = (
+            parts[1].strip().replace("-", "_") if len(parts) > 1 else "all"
+        )  # Replace hyphens with underscores
 
-            # Return the transformed values
-            return {"name": first_part, "subject": second_part}
+        # Return the transformed values
+        return {"name": first_part, "subject": second_part}
 
-        bench_dataset = load_dataset(
-            "pharaouk/dharma-1",
-            data_files={
-                "eval": "dharma_1_mini.json",
-            },
-        )
-        bench_dataset["eval"] = bench_dataset["eval"].map(transform_subject)
-    elif trainer.args.bench_dataset == "mmlu-zs":
+    if trainer.args.bench_dataset == "mmlu-zs":
         bench_dataset = load_dataset(
             "openaccess-ai-collective/mmlu-evals",
             data_files={
@@ -165,6 +157,17 @@ def bench_eval_callback_factory(trainer, tokenizer):
             },
         )
         # bench_dataset = bench_dataset.remove_columns('subject')
+    elif "/" in trainer.args.bench_dataset:
+        bench_ds = trainer.args.bench_dataset
+        bench_ds_name = "/".join(bench_ds.split("/", 2)[:2])
+        bench_ds_data_file = "/".join(bench_ds.split("/", 2)[2:])
+        bench_dataset = load_dataset(
+            bench_ds_name,
+            data_files={
+                "eval": bench_ds_data_file,
+            },
+        )
+        bench_dataset["eval"] = bench_dataset["eval"].map(transform_bench_subject)
     else:
         raise ValueError(
             f"unhandled value `{trainer.args.bench_dataset}` for bench_dataset training args"

src/axolotl/utils/trainer.py

@@ -138,9 +138,9 @@ class AxolotlTrainingArguments(TrainingArguments):
         default="eval", metadata={"help": "The benchmark split to run on"}
     )
     bench_dataset: Optional[str] = field(
-        default="sampled",
+        default="pharaouk/dharma-1/dharma_1_mini.json",
         metadata={
-            "help": "Benchmark dataset to use: options are `mmlu-zs`, `mmlu-fs`, `sampled`"
+            "help": "Benchmark dataset to use: options are `mmlu-zs`, `mmlu-fs`, or the full path to the dataset file"
         },
     )
     do_bench_eval: Optional[bool] = field(