Add files using upload-large-folder tool

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_105.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalae_splitae_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer105/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalae_splitae_splitab \
+ --output-prefix $FINAL_DIR/tokenizer105/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_108.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalae_splitae_splitae
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer108/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalae_splitae_splitae \
+ --output-prefix $FINAL_DIR/tokenizer108/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_122.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaf_splitae
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer122/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaf_splitae \
+ --output-prefix $FINAL_DIR/tokenizer122/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_126.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalag_splitab_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer126/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalag_splitab_splitab \
+ --output-prefix $FINAL_DIR/tokenizer126/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_136.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalag_splitad_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer136/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalag_splitad_splitab \
+ --output-prefix $FINAL_DIR/tokenizer136/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_167.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalai_splitab_splitac
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer167/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalai_splitab_splitac \
+ --output-prefix $FINAL_DIR/tokenizer167/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_176.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalai_splitad_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer176/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalai_splitad_splitab \
+ --output-prefix $FINAL_DIR/tokenizer176/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_18.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalab_splitaa_splitaa
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer18/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalab_splitaa_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer18/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_183.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalai_splitae_splitad
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer183/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalai_splitae_splitad \
+ --output-prefix $FINAL_DIR/tokenizer183/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_19.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalab_splitaa_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer19/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalab_splitaa_splitab \
+ --output-prefix $FINAL_DIR/tokenizer19/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_197.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaj_splitac_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer197/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaj_splitac_splitab \
+ --output-prefix $FINAL_DIR/tokenizer197/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_210.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalak_splitaa_splitac
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer210/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalak_splitaa_splitac \
+ --output-prefix $FINAL_DIR/tokenizer210/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_217.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalak_splitac_splitad
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer217/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalak_splitac_splitad \
+ --output-prefix $FINAL_DIR/tokenizer217/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_218.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalak_splitac_splitae
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer218/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalak_splitac_splitae \
+ --output-prefix $FINAL_DIR/tokenizer218/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_24.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalab_splitab_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer24/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalab_splitab_splitab \
+ --output-prefix $FINAL_DIR/tokenizer24/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_27.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalab_splitab_splitae
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer27/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalab_splitab_splitae \
+ --output-prefix $FINAL_DIR/tokenizer27/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_276.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalan_splitac_splitaa
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer276/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalan_splitac_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer276/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_278.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalan_splitac_splitac
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer278/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalan_splitac_splitac \
+ --output-prefix $FINAL_DIR/tokenizer278/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_288.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalao_splitaa_splitaa
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer288/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalao_splitaa_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer288/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_300.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalao_splitad_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer300/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalao_splitad_splitab \
+ --output-prefix $FINAL_DIR/tokenizer300/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_304.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalao_splitae_splitaa
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer304/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalao_splitae_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer304/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_317.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalap_splitab_splitac
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer317/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalap_splitab_splitac \
+ --output-prefix $FINAL_DIR/tokenizer317/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_361.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalar_splitaa_splitad
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer361/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalar_splitaa_splitad \
+ --output-prefix $FINAL_DIR/tokenizer361/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_402.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalas_splitae_splitac
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer402/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalas_splitae_splitac \
+ --output-prefix $FINAL_DIR/tokenizer402/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_41.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalac_splitaa_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer41/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalac_splitaa_splitab \
+ --output-prefix $FINAL_DIR/tokenizer41/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_411.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalat_splitad_splitac
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer411/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalat_splitad_splitac \
+ --output-prefix $FINAL_DIR/tokenizer411/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_445.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalav_splitaa_splitad
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer445/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalav_splitaa_splitad \
+ --output-prefix $FINAL_DIR/tokenizer445/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_458.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalav_splitad_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer458/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalav_splitad_splitab \
+ --output-prefix $FINAL_DIR/tokenizer458/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_470.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaw_splitaa_splitac
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer470/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaw_splitaa_splitac \
+ --output-prefix $FINAL_DIR/tokenizer470/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_472.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaw_splitaa_splitae
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer472/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaw_splitaa_splitae \
+ --output-prefix $FINAL_DIR/tokenizer472/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_474.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaw_splitab_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer474/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaw_splitab_splitab \
+ --output-prefix $FINAL_DIR/tokenizer474/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_476.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaw_splitab_splitad
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer476/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaw_splitab_splitad \
+ --output-prefix $FINAL_DIR/tokenizer476/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_491.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaw_splitae_splitad
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer491/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaw_splitae_splitad \
+ --output-prefix $FINAL_DIR/tokenizer491/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_496.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalax_splitac_splitaa
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer496/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalax_splitac_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer496/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_499.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalax_splitac_splitad
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer499/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalax_splitac_splitad \
+ --output-prefix $FINAL_DIR/tokenizer499/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_501.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalax_splitad
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer501/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalax_splitad \
+ --output-prefix $FINAL_DIR/tokenizer501/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_52.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalac_splitac_splitac
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer52/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalac_splitac_splitac \
+ --output-prefix $FINAL_DIR/tokenizer52/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_533.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaz_splitaa_splitad
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer533/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaz_splitaa_splitad \
+ --output-prefix $FINAL_DIR/tokenizer533/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_540.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaz_splitac
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer540/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalaz_splitac \
+ --output-prefix $FINAL_DIR/tokenizer540/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_553.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalba_splitaa_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer553/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalba_splitaa_splitab \
+ --output-prefix $FINAL_DIR/tokenizer553/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_562.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalba_splitac_splitaa
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer562/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalba_splitac_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer562/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_573.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalba_splitaf
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer573/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalba_splitaf \
+ --output-prefix $FINAL_DIR/tokenizer573/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_574.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbb_splitaa
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer574/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbb_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer574/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_583.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbb_splitac_splitad
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer583/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbb_splitac_splitad \
+ --output-prefix $FINAL_DIR/tokenizer583/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_584.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbb_splitac_splitae
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer584/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbb_splitac_splitae \
+ --output-prefix $FINAL_DIR/tokenizer584/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_592.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbb_splitae_splitac
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer592/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbb_splitae_splitac \
+ --output-prefix $FINAL_DIR/tokenizer592/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_596.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbc_splitaa_splitaa
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer596/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbc_splitaa_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer596/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_629.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbd_splitac_splitab
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer629/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbd_splitac_splitab \
+ --output-prefix $FINAL_DIR/tokenizer629/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_630.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbd_splitac_splitac
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer630/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbd_splitac_splitac \
+ --output-prefix $FINAL_DIR/tokenizer630/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_635.sh

@@ -0,0 +1,25 @@
+#!/bin/bash
+set -m
+sed -i -e "s/raw_content/text/g" /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbd_splitaf
+echo "above deepspeed$"
+FINAL_DIR=/mnt/weka/peacock/idc/hineng/tok_en-bn-hn/
+#TOKENIZER=facebook/nllb-200-distilled-600M
+#TOKENIZER_TYPE=HuggingFaceTokenizer
+TOKENIZER_PATH=/mnt/weka/peacock/tokenization/trained-tokenizer/enhiben_50k
+TOKENIZER=$TOKENIZER_PATH/all.model
+VOCAB_FILE=$TOKENIZER_PATH/all.vocab
+TOKENIZER_TYPE=SentencePieceTokenizer
+mkdir -p $FINAL_DIR/tokenizer635/
+cd /mnt/weka/peacock/experiments/llama/Megatron-DeepSpeed
+echo "inside deepspeed"
+pwd
+python3 tools/preprocess_data.py \
+ --input /mnt/weka/peacock/idc/datasets/hn_eng_bn/tok_en-bn-hn/split//finalbd_splitaf \
+ --output-prefix $FINAL_DIR/tokenizer635/ \
+  --tokenizer-model $TOKENIZER \
+  --vocab-file $VOCAB_FILE  \
+    --dataset-impl mmap --tokenizer-type $TOKENIZER_TYPE\
+    --append-eod --workers 8 # --partitions 16 #--chunk-size 50   
+
+  # --merge-file $MERGES_FILE \
+