Update README.md

README.md

@@ -44,7 +44,7 @@ python3 quantize_quark.py --model_dir $MODEL_DIR \
                           --skip_evaluation \
                           --multi_gpu \
                           --model_export hf_format \
-                          --output_dir amd/DeepSeek-R1-0528-MXFP4-Preview
+                          --output_dir amd/DeepSeek-R1-0528-MXFP4
 ```
 
 # Deployment
@@ -63,7 +63,7 @@ The model was evaluated on AIME24, GPQA Diamond, and MATH-500 benchmarks using t
    </td>
    <td><strong>DeepSeek-R1-0528 </strong>
    </td>
-   <td><strong>DeepSeek-R1-0528-MXFP4-Preview(this model)</strong>
+   <td><strong>DeepSeek-R1-0528-MXFP4 (this model)</strong>
    </td>
    <td><strong>Recovery</strong>
    </td>
@@ -110,14 +110,14 @@ The results of AIME24, MATH-500, and GPQA Diamond, were obtained using forked [l
 export VLLM_QUARK_F4F6_OFFLINE_DEQUANT_TMPENVVAR=1
 
 # Set model args
-OUTPUT_DIR="results/DeepSeek-R1-0528-MXFP4-Preview-Seed"
+OUTPUT_DIR="results/DeepSeek-R1-0528-MXFP4-Seed"
 LOG="logs/deepseek_0528_maxfp4.log"
 
 # Evaluating 10 rounds 
 for i in $(seq 1 10); do
     # seed in [0, 2**30 - 1]
     SEED=$(shuf -i 0-1073741823 -n 1)
-    MODEL_ARGS="model_name=amd/DeepSeek-R1-0528-MXFP4-Preview,dtype=bfloat16,tensor_parallel_size=8,max_model_length=71536,max_num_batched_tokens=32768,gpu_memory_utilization=0.85,generation_parameters={max_new_tokens:65536,temperature:0.6,top_p:0.95,seed:$SEED}"
+    MODEL_ARGS="model_name=amd/DeepSeek-R1-0528-MXFP4,dtype=bfloat16,tensor_parallel_size=8,max_model_length=71536,max_num_batched_tokens=32768,gpu_memory_utilization=0.85,generation_parameters={max_new_tokens:65536,temperature:0.6,top_p:0.95,seed:$SEED}"
 
     lighteval vllm $MODEL_ARGS "custom|aime24_single|0|0,custom|math_500_single|0|0,custom|gpqa:diamond_single|0|0" \
         --use-chat-template \