GenVRadmin commited on
Commit
f47005f
·
verified ·
1 Parent(s): d748c03

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -6,7 +6,7 @@ This model is finetuned from HuggingFaceH4/zephyr-7b-gemma-v0.1 and is finetuned
6
  To improve the resoning and maths skills, we first SFT tune the gemma on Microsoft's Orca datasets.
7
 
8
  We utilize Orca maths Hindi dataset: GenVRadmin/Aryabhatta-Orca-Maths-Hindi \
9
- And original Orca maths dataset: microsoft/orca-math-word-problems-200k \
10
 
11
  This pushes the MATHS score from 24.3 in Gemma-7B to 25.5 in Zephyr-Gemma and 31.6 in GemmaOrca.
12
 
@@ -17,7 +17,7 @@ This is then finetuned on various open sourced datasets like:
17
  Telugu-LLM-Labs/yahma_alpaca_cleaned_telugu_filtered_and_romanized \
18
  Telugu-LLM-Labs/teknium_GPTeacher_general_instruct_telugu_filtered_and_romanized \
19
  abhinand/tamil-alpaca \
20
- Tensoic/airoboros-3.2_kn \
21
  Tensoic/gpt-teacher_kn \
22
  Tensoic/Alpaca-Gujarati \
23
  HydraIndicLM/bengali_alpaca_dolly_67k \
@@ -26,7 +26,7 @@ pankajmathur/alpaca_orca \
26
  OdiaGenAI/Odia_Alpaca_instructions_52k \
27
  OdiaGenAI/gpt-teacher-roleplay-odia-3k \
28
  GenVRadmin/Samvaad-Punjabi-Mini \
29
- pankajmathur/WizardLM_Orca \
30
 
31
  The model achieves following scores on benchmarks:
32
 
 
6
  To improve the resoning and maths skills, we first SFT tune the gemma on Microsoft's Orca datasets.
7
 
8
  We utilize Orca maths Hindi dataset: GenVRadmin/Aryabhatta-Orca-Maths-Hindi \
9
+ And original Orca maths dataset: microsoft/orca-math-word-problems-200k
10
 
11
  This pushes the MATHS score from 24.3 in Gemma-7B to 25.5 in Zephyr-Gemma and 31.6 in GemmaOrca.
12
 
 
17
  Telugu-LLM-Labs/yahma_alpaca_cleaned_telugu_filtered_and_romanized \
18
  Telugu-LLM-Labs/teknium_GPTeacher_general_instruct_telugu_filtered_and_romanized \
19
  abhinand/tamil-alpaca \
20
+ Tensoic/airoboros-3.2_kn \
21
  Tensoic/gpt-teacher_kn \
22
  Tensoic/Alpaca-Gujarati \
23
  HydraIndicLM/bengali_alpaca_dolly_67k \
 
26
  OdiaGenAI/Odia_Alpaca_instructions_52k \
27
  OdiaGenAI/gpt-teacher-roleplay-odia-3k \
28
  GenVRadmin/Samvaad-Punjabi-Mini \
29
+ pankajmathur/WizardLM_Orca
30
 
31
  The model achieves following scores on benchmarks:
32