Rhett1 commited on
Commit
a0d33a3
·
verified ·
1 Parent(s): 54551f8

Complete the LLaMA-2-13B base block: TR-3 HellaSwag 10-shot (77.04)

Browse files
Files changed (1) hide show
  1. README.md +7 -7
README.md CHANGED
@@ -138,14 +138,14 @@ Perplexity is **token-level** over fixed non-overlapping 2048-token windows, fol
138
  | | ASVD | 41.43 | 68.01 | 29.71 | 7.51 |
139
  | **LLaMA-2-7B (base)** | *dense FP16* | 45.97 | 78.95 | 64.39 | 13.12 |
140
  | | TR-3 (ours) | 39.49 | 71.11 | 39.75 | 6.52 |
141
- | | SVD-LLM | -- | -- | -- | -- |
142
- | | SVD-LLM V2 ‡ | -- | -- | -- | -- |
143
- | | ASVD | -- | -- | -- | -- |
144
  | **LLaMA-2-13B (base)** | *dense FP16* | 55.54 | 82.14 | 70.69 | 23.28 |
145
- | | TR-3 (ours) | 50.46 | -- | 50.96 | 14.33 |
146
- | | ASVD | -- | -- | -- | -- |
147
- | | SVD-LLM | -- | -- | -- | -- |
148
- | | SVD-LLM V2 ‡ | -- | -- | -- | -- |
149
  | **Qwen2.5-3B** | *dense FP16* | 66.42 | 75.26 | 45.78 | 10.31 |
150
  | | TR-3 (ours) | 56.02 | 66.39 | 27.61 | 41.32 |
151
  | | TT-3 (open chain) | 41.97 | 51.65 | 11.86 | 9.78 |
 
138
  | | ASVD | 41.43 | 68.01 | 29.71 | 7.51 |
139
  | **LLaMA-2-7B (base)** | *dense FP16* | 45.97 | 78.95 | 64.39 | 13.12 |
140
  | | TR-3 (ours) | 39.49 | 71.11 | 39.75 | 6.52 |
141
+ | | SVD-LLM | 32.74 | 58.36 | 16.43 | 1.59 |
142
+ | | SVD-LLM V2 ‡ | 34.33 | 59.36 | 19.34 | 1.97 |
143
+ | | ASVD | 26.58 | 61.21 | 20.69 | 1.29 |
144
  | **LLaMA-2-13B (base)** | *dense FP16* | 55.54 | 82.14 | 70.69 | 23.28 |
145
+ | | TR-3 (ours) | 50.46 | 77.04 | 50.96 | 14.33 |
146
+ | | ASVD | 42.44 | 70.15 | 35.50 | 5.23 |
147
+ | | SVD-LLM | 42.70 | 63.80 | 21.87 | 4.25 |
148
+ | | SVD-LLM V2 ‡ | 44.67 | 65.00 | 24.49 | 4.93 |
149
  | **Qwen2.5-3B** | *dense FP16* | 66.42 | 75.26 | 45.78 | 10.31 |
150
  | | TR-3 (ours) | 56.02 | 66.39 | 27.61 | 41.32 |
151
  | | TT-3 (open chain) | 41.97 | 51.65 | 11.86 | 9.78 |