From 05ce8eb1e6f5a844d2592ecb73e1cf8e3b870710 Mon Sep 17 00:00:00 2001
From: Juhan Bae <pomonam15@gmail.com>
Date: Wed, 26 Jun 2024 02:57:36 -0400
Subject: [PATCH] minor

---
 examples/glue/README.md  | 9 +++++----
 examples/glue/analyze.py | 2 +-
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/examples/glue/README.md b/examples/glue/README.md
index 9b4671a..ac6e182 100644
--- a/examples/glue/README.md
+++ b/examples/glue/README.md
@@ -63,7 +63,7 @@ Can we remove top positively influential training examples to make some queries
 selects correctly classified query data point, removes top-k positively influential training samples, and retrain the network with the modified dataset to see if that query 
 data point gets misclassified.
 
-We first need to compute pairwise influence scores for the `RTE` dataset:
+We first need to compute pairwise influence scores for the `RTE` dataset (A6000 GPU was used to run these experiments):
 
 ```bash
 python train.py --dataset_name rte \
@@ -76,13 +76,13 @@ python train.py --dataset_name rte \
     --seed 1004
 
 python analyze.py --dataset_name rte \
-    --query_batch_size 175 \
+    --query_batch_size 70 \
     --train_batch_size 128 \
     --checkpoint_dir ./checkpoints \
     --factor_strategy ekfac
     
 python analyze.py --dataset_name rte \
-    --query_batch_size 175 \
+    --query_batch_size 139 \
     --train_batch_size 128 \
     --checkpoint_dir ./checkpoints \
     --factor_strategy identity
@@ -97,7 +97,8 @@ python analyze.py --dataset_name rte \
 ## Evaluating Linear Datamodeling Score
 
 The `evaluate_lds.py` script computes the [linear datamodeling score (LDS)](https://arxiv.org/abs/2303.14186). It measures the LDS obtained by 
-retraining the network 500 times with different subsets of the dataset (5 repeats and 100 masks). By running `evaludate_lds.py`, we obtain `xx` LDS (we get `xx` LDS with the half precision).
+retraining the network 500 times with different subsets of the dataset (5 repeats and 100 masks). 
+By running `evaludate_lds.py`, we obtain `xx` LDS (we get `xx` LDS with the half precision).
 
 The script also includes functionality to print out top influential sequences for a given query.
 
diff --git a/examples/glue/analyze.py b/examples/glue/analyze.py
index df610f5..a824d5d 100644
--- a/examples/glue/analyze.py
+++ b/examples/glue/analyze.py
@@ -176,7 +176,7 @@ def main():
         dataset=train_dataset,
         per_device_batch_size=None,
         factor_args=factor_args,
-        overwrite_output_dir=True,
+        overwrite_output_dir=False,
         initial_per_device_batch_size_attempt=512,
     )