Update README.md

README.md

@@ -130,10 +130,11 @@ model.save_pretrained("Meta-Llama-3.1-70B-Instruct-quantized.w8a16")
 
 The model was evaluated on MMLU, ARC-Challenge, GSM-8K, Hellaswag, Winogrande and TruthfulQA.
 Evaluation was conducted using the Neural Magic fork of [lm-evaluation-harness](https://github.com/neuralmagic/lm-evaluation-harness/tree/llama_3.1_instruct) (branch llama_3.1_instruct) and the [vLLM](https://docs.vllm.ai/en/stable/) engine.
-This version of the lm-evaluation-harness includes versions of ARC-Challenge and GSM-8K that match the prompting style of [Meta-Llama-3.1-Instruct-evals](https://huggingface.co/datasets/meta-llama/Meta-Llama-3.1-70B-Instruct-evals).
+This version of the lm-evaluation-harness includes versions of MMLU, ARC-Challenge and GSM-8K that match the prompting style of [Meta-Llama-3.1-Instruct-evals](https://huggingface.co/datasets/meta-llama/Meta-Llama-3.1-70B-Instruct-evals).
 
 ### Accuracy
 
+#### Open LLM Leaderboard evaluation scores
 <table>
   <tr>
    <td><strong>Benchmark</strong>
@@ -148,21 +149,31 @@ This version of the lm-evaluation-harness includes versions of ARC-Challenge and
   <tr>
    <td>MMLU (5-shot)
    </td>
-   <td>82.21
+   <td>83.88
    </td>
-   <td>82.12
+   <td>81.07
    </td>
-   <td>99.9%
+   <td>96.6%
+   </td>
+  </tr>
+  <tr>
+   <td>MMLU (CoT, 0-shot)
+   </td>
+   <td>85.74
+   </td>
+   <td>83.29
+   </td>
+   <td>97.1%
    </td>
   </tr>
   <tr>
    <td>ARC Challenge (0-shot)
    </td>
-   <td>95.05
+   <td>93.26
    </td>
-   <td>93.60
+   <td>91.98
    </td>
-   <td>98.5%
+   <td>98.6%
    </td>
   </tr>
   <tr>
@@ -208,11 +219,11 @@ This version of the lm-evaluation-harness includes versions of ARC-Challenge and
   <tr>
    <td><strong>Average</strong>
    </td>
-   <td><strong>83.60</strong>
+   <td><strong>83.89</strong>
    </td>
-   <td><strong>82.66</strong>
+   <td><strong>82.54</strong>
    </td>
-   <td><strong>99.1%</strong>
+   <td><strong>98.4%</strong>
    </td>
   </tr>
 </table>
@@ -225,17 +236,30 @@ The results were obtained using the following commands:
 ```
 lm_eval \
   --model vllm \
-  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=4 \
-  --tasks mmlu \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=3850,max_gen_toks=10,tensor_parallel_size=1 \
+  --tasks mmlu_llama_3.1_instruct \
+  --fewshot_as_multiturn \
+  --apply_chat_template \
   --num_fewshot 5 \
   --batch_size auto
 ```
 
+#### MMLU-CoT
+```
+lm_eval \
+  --model vllm \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=4064,max_gen_toks=1024,tensor_parallel_size=1 \
+  --tasks mmlu_cot_0shot_llama_3.1_instruct \
+  --apply_chat_template \
+  --num_fewshot 0 \
+  --batch_size auto
+```
+
 #### ARC-Challenge
 ```
 lm_eval \
   --model vllm \
-  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=4 \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=3940,max_gen_toks=100,tensor_parallel_size=1 \
   --tasks arc_challenge_llama_3.1_instruct \
   --apply_chat_template \
   --num_fewshot 0 \
@@ -246,7 +270,7 @@ lm_eval \
 ```
 lm_eval \
   --model vllm \
-  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=4 \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=4096,max_gen_toks=1024,tensor_parallel_size=1 \
   --tasks gsm8k_cot_llama_3.1_instruct \
   --fewshot_as_multiturn \
   --apply_chat_template \
@@ -258,7 +282,7 @@ lm_eval \
 ```
 lm_eval \
   --model vllm \
-  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=4 \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=1 \
   --tasks hellaswag \
   --num_fewshot 10 \
   --batch_size auto
@@ -268,7 +292,7 @@ lm_eval \
 ```
 lm_eval \
   --model vllm \
-  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=4 \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=1 \
   --tasks winogrande \
   --num_fewshot 5 \
   --batch_size auto
@@ -278,7 +302,7 @@ lm_eval \
 ```
 lm_eval \
   --model vllm \
-  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=4 \
+  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-Instruct-quantized.w8a16",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=1 \
   --tasks truthfulqa \
   --num_fewshot 0 \
   --batch_size auto