GPTQ add Arkiv link, move file location (#1100)

## Purpose ##
* Better docstring for GPTQ
* Reduce unnecessary file hierarchy

Signed-off-by: Kyle Sayers <[email protected]>

Author: kylesayrs

src/llmcompressor/modifiers/quantization/gptq/base.py

@@ -16,7 +16,7 @@
 from llmcompressor.core import State
 from llmcompressor.modifiers import Modifier, ModifierFactory
 from llmcompressor.modifiers.quantization.calibration import freeze_module_quantization
-from llmcompressor.modifiers.quantization.gptq.utils.gptq_quantize import (
+from llmcompressor.modifiers.quantization.gptq.gptq_quantize import (
     accumulate_hessian,
     make_empty_hessian,
     quantize_weight,
@@ -36,7 +36,9 @@
 
 class GPTQModifier(Modifier, HooksMixin):
     """
-    Modifier for applying the one-shot OBCQ algorithm to a model
+    Implements the GPTQ algorithm from https://arxiv.org/abs/2210.17323. This modifier
+    uses activations to calibrate a hessian matrix, which is then used to determine
+    optimal quantizion values and orderings for the model weights.
 
     | Sample yaml:
     | test_stage: