Add files using upload-large-folder tool

hn_eng_bn/tok_en-bn-hn/tok_files/tokenizer_104.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_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/tokenizer104/
+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_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer104/ \
+  --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_112.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_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/tokenizer112/
+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_splitaa_splitac \
+ --output-prefix $FINAL_DIR/tokenizer112/ \
+  --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_113.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_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/tokenizer113/
+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_splitaa_splitad \
+ --output-prefix $FINAL_DIR/tokenizer113/ \
+  --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_134.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_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/tokenizer134/
+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_splitac_splitae \
+ --output-prefix $FINAL_DIR/tokenizer134/ \
+  --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_14.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//finalaa_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/tokenizer14/
+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//finalaa_splitac_splitae \
+ --output-prefix $FINAL_DIR/tokenizer14/ \
+  --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_140.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_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/tokenizer140/
+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_splitae_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer140/ \
+  --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_149.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//finalah_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/tokenizer149/
+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//finalah_splitab_splitac \
+ --output-prefix $FINAL_DIR/tokenizer149/ \
+  --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_15.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//finalaa_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/tokenizer15/
+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//finalaa_splitad \
+ --output-prefix $FINAL_DIR/tokenizer15/ \
+  --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_153.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//finalah_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/tokenizer153/
+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//finalah_splitac_splitab \
+ --output-prefix $FINAL_DIR/tokenizer153/ \
+  --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_16.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//finalaa_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/tokenizer16/
+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//finalaa_splitae \
+ --output-prefix $FINAL_DIR/tokenizer16/ \
+  --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_163.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//finalah_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/tokenizer163/
+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//finalah_splitaf \
+ --output-prefix $FINAL_DIR/tokenizer163/ \
+  --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_174.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_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/tokenizer174/
+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_splitac_splitae \
+ --output-prefix $FINAL_DIR/tokenizer174/ \
+  --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_179.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_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/tokenizer179/
+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_splitae \
+ --output-prefix $FINAL_DIR/tokenizer179/ \
+  --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_189.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_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/tokenizer189/
+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_splitaa_splitad \
+ --output-prefix $FINAL_DIR/tokenizer189/ \
+  --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_209.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_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/tokenizer209/
+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_splitab \
+ --output-prefix $FINAL_DIR/tokenizer209/ \
+  --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_213.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_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/tokenizer213/
+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_splitab \
+ --output-prefix $FINAL_DIR/tokenizer213/ \
+  --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_223.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_splitad_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/tokenizer223/
+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_splitad_splitae \
+ --output-prefix $FINAL_DIR/tokenizer223/ \
+  --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_263.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//finalam_splitad_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/tokenizer263/
+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//finalam_splitad_splitae \
+ --output-prefix $FINAL_DIR/tokenizer263/ \
+  --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_273.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_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/tokenizer273/
+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_splitab_splitac \
+ --output-prefix $FINAL_DIR/tokenizer273/ \
+  --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_280.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_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/tokenizer280/
+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_splitae \
+ --output-prefix $FINAL_DIR/tokenizer280/ \
+  --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_283.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_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/tokenizer283/
+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_splitad_splitac \
+ --output-prefix $FINAL_DIR/tokenizer283/ \
+  --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_286.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_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/tokenizer286/
+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_splitae \
+ --output-prefix $FINAL_DIR/tokenizer286/ \
+  --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_287.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_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/tokenizer287/
+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_splitaf \
+ --output-prefix $FINAL_DIR/tokenizer287/ \
+  --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_290.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_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/tokenizer290/
+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_splitac \
+ --output-prefix $FINAL_DIR/tokenizer290/ \
+  --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_294.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_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/tokenizer294/
+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_splitab_splitab \
+ --output-prefix $FINAL_DIR/tokenizer294/ \
+  --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_316.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_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/tokenizer316/
+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_splitab \
+ --output-prefix $FINAL_DIR/tokenizer316/ \
+  --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_326.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_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/tokenizer326/
+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_splitad_splitab \
+ --output-prefix $FINAL_DIR/tokenizer326/ \
+  --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_337.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//finalaq_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/tokenizer337/
+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//finalaq_splitaa_splitab \
+ --output-prefix $FINAL_DIR/tokenizer337/ \
+  --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_338.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//finalaq_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/tokenizer338/
+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//finalaq_splitaa_splitac \
+ --output-prefix $FINAL_DIR/tokenizer338/ \
+  --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_366.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_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/tokenizer366/
+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_splitab_splitad \
+ --output-prefix $FINAL_DIR/tokenizer366/ \
+  --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_37.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_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/tokenizer37/
+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_splitae_splitad \
+ --output-prefix $FINAL_DIR/tokenizer37/ \
+  --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_389.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_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/tokenizer389/
+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_splitab_splitae \
+ --output-prefix $FINAL_DIR/tokenizer389/ \
+  --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_403.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_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/tokenizer403/
+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_splitad \
+ --output-prefix $FINAL_DIR/tokenizer403/ \
+  --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_42.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_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/tokenizer42/
+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_splitac \
+ --output-prefix $FINAL_DIR/tokenizer42/ \
+  --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_426.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//finalau_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/tokenizer426/
+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//finalau_splitac_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer426/ \
+  --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_431.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//finalau_splitad_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/tokenizer431/
+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//finalau_splitad_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer431/ \
+  --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_436.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//finalau_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/tokenizer436/
+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//finalau_splitae_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer436/ \
+  --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_441.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//finalau_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/tokenizer441/
+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//finalau_splitaf \
+ --output-prefix $FINAL_DIR/tokenizer441/ \
+  --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_485.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_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/tokenizer485/
+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_splitad_splitac \
+ --output-prefix $FINAL_DIR/tokenizer485/ \
+  --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_49.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_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/tokenizer49/
+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_splitab_splitae \
+ --output-prefix $FINAL_DIR/tokenizer49/ \
+  --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_500.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_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/tokenizer500/
+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_splitae \
+ --output-prefix $FINAL_DIR/tokenizer500/ \
+  --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_504.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_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/tokenizer504/
+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_splitae_splitac \
+ --output-prefix $FINAL_DIR/tokenizer504/ \
+  --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_505.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_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/tokenizer505/
+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_splitae_splitad \
+ --output-prefix $FINAL_DIR/tokenizer505/ \
+  --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_507.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_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/tokenizer507/
+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_splitaf \
+ --output-prefix $FINAL_DIR/tokenizer507/ \
+  --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_510.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//finalay_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/tokenizer510/
+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//finalay_splitab_splitab \
+ --output-prefix $FINAL_DIR/tokenizer510/ \
+  --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_527.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//finalay_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/tokenizer527/
+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//finalay_splitae_splitad \
+ --output-prefix $FINAL_DIR/tokenizer527/ \
+  --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_55.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_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/tokenizer55/
+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_splitad \
+ --output-prefix $FINAL_DIR/tokenizer55/ \
+  --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_551.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_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/tokenizer551/
+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_splitaf \
+ --output-prefix $FINAL_DIR/tokenizer551/ \
+  --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_56.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_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/tokenizer56/
+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_splitae \
+ --output-prefix $FINAL_DIR/tokenizer56/ \
+  --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_580.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_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/tokenizer580/
+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_splitaa \
+ --output-prefix $FINAL_DIR/tokenizer580/ \
+  --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 \
+