Share your thoughts, 1 month free Claude Pro on usSee more

Reasoning on WinoGrande

85.2Accuracy

InternLM2-20B

Updated 12d ago

Evaluation Results

Method
InternLM2-20B 2024.03	85.2	-
InternLM2-7B 2024.03	84.7	-
Qwen3.5-9B 2026.04	82.64	-
Llama-3.3-70B-Instruct 2026.04	81.06	-
Mixtral-8x7B-v0.1 2024.03	77.3	-
InternLM2-20B-Base 2024.03	76.2	-
Qwen3-14B 2026.04	76.16	-
Mistral-7B-v0.1 2024.03	75.3	-
InternLM2-Chat-20B-SFT 2024.03	75	-
Phi-mini (teacher) 2026.05	74.9	63.72
InternLM2-Chat-20B 2024.03	74.8	-
InternLM2-7B-Base 2024.03	74.6	-
AWQ 2025.03	74	-
Hybrid-Prefix 2026.04	74	-
XekRung-8B 2026.04	73.24	-
LLAMA-3-8B 2026.04	73.16	-
ClusComp 2025.03	72.9	-
Llama-3-8B 2025.03	72.8	-
SliM-LLM 2025.03	72.8	-
GPTQ 2025.03	72.6	-
SliM-LLM 2025.03	72.6	-
QuIP 2025.03	72.5	-
SecGPT-14B 2026.04	72.45	-
Llama2-13B 2024.03	72.3	-
Qwen-4B (teacher) 2026.05	72.3	66.42
AWQ 2025.03	72.1	-
GPTQ 2025.03	71.1	-
RTN 2025.03	71	-
ClusComp 2025.03	71	-
Llama-3B (teacher) 2026.05	70.01	46.93
Baichuan2-13B-Base 2024.03	70	-
Original 2026.02	70	-
GPTQ 2026.02	70	-
SVD-LLM V2 2026.05	70	-
Baseline 2026.04	69.53	-
Llama2-7B 2024.03	69.5	-
Hybrid-Prompt 2026.04	68.9	-
Baseline 2026.04	68.82	-
Qwen3-8B 2026.04	68.75	-
Baichuan2-7B-Base 2024.03	68.7	-
GPT-3.5 2024.03	68.7	-
Qwen-7B 2024.03	68.6	-
FO-LoRA 2026.04	68.5	-
SPQ 2026.02	68	-
Baseline 2026.04	67.64	-
Qwen-14B 2024.03	67.5	-
FO-Prompt 2026.04	67.2	-
LLAMA-2-7B 2026.04	67.17	-
ClusComp 2025.03	67.1	-
LLAMA-1-7B 2026.04	67.01	-
Baseline 2026.05	67	-
SAES-SVD + PARSE 2026.05	67	-
FO-LoRA 2026.04	66.7	-
Hybrid-LoRA 2026.04	66.7	-
Hybrid-LoRA 2026.04	66.3	-
Hybrid-Prompt 2026.04	66.3	-
FO-Prefix 2026.04	66.2	-
ASVD 2026.02	66	-
SAES-SVD + PARSE 2026.05	66	-
Phi-3 2025.05	65.98	-
Arrow 2025.05	65.98	-
MoBiE 2026.04	65.9	-
InternLM2-Chat-7B-SFT 2024.03	65.8	-
InternLM2-Chat-7B 2024.03	65.8	-
FO-Prompt 2026.04	65.8	-
GPTQ 2026.04	65.48	-
SparseGPT 2026.02	65	-
NoWag 2026.04	64.96	-
GenKnowSub 2025.05	64.72	-
AWQ 2025.03	64.7	-
GenKnowSub 2025.05	64.64	-
GenKnowSub 2025.05	64.4	-
Hybrid-Prefix 2026.04	64.3	-
FO-Prefix 2026.04	64.1	-
X-Token 2026.05	63.61	40.15
X-Token (P-KL) 2026.05	63.46	38.85
GenKnowSub 2025.05	63.3	-
X-Token 2026.05	63.3	40.48
SAES-SVD + PARSE 2026.05	63	-
GOLD 2026.05	62.95	35.03
X-Token 2026.05	62.74	38.49
Llama-3B → 1B 2026.05	62.7	38.4
GOLD 2026.05	62.6	38.66
ULD 2026.05	62.59	38.31
ClusComp 2025.03	62.4	-
ULD 2026.05	61.96	36.77
X-Token (H-KL) 2026.05	61.87	39.18
LBLLM 2026.04	61.72	-
ChatGLM3-6B 2024.03	61.7	-
Continued pre-training 2026.05	61.6	36.63
Llama-1B (base) 2026.05	61.48	33.96
Mixtral-8x7B-Instruct-v0.1 2024.03	60.9	-
GPTQ 2025.03	60.9	-
SmolLM 2 2025.06	60.1	-
Hybrid-Prefix 2026.04	60	-
Hybrid-Prompt 2026.04	59.9	-
Qwen 2.5 2025.06	59.4	-
Qwen 2 2025.06	59.2	-
Stable LM 2 2025.06	59.2	-
FO-LoRA 2026.04	59	-

Showing 100 of 168 rows