Narrative Generation on TinyStories 21 (test)
1.62Speedup (x)SWIFT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SWIFTModel=LLaMA-2-70B, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 1.62 | 99 | |
| SDFPModel=LLaMA-2-70B, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 1.55 | 95 | |
| SWIFTModel=LLaMA-2-13B, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 1.53 | 100 | |
| SWIFTModel=LLaMA-2-13B-Chat, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 1.5 | 99 | |
| SDFPModel=LLaMA-2-13B, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 1.49 | 98 | |
| SDFPModel=LLaMA-2-13B-Chat, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 1.44 | 95 | |
| Lookahead DecodingModel=LLaMA-2-13B-Chat, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 1.4 | — | |
| Lookahead DecodingModel=LLaMA-2-13B, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 1.35 | — | |
| Lookahead DecodingModel=LLaMA-2-70B, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 1.35 | — | |
| VANILLAModel=LLaMA-2-13B, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 1 | — | |
| VANILLAModel=LLaMA-2-13B-Chat, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 1 | — | |
| VANILLAModel=LLaMA-2-70B, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 1 | — | |
| Parallel DecodingModel=LLaMA-2-13B-Chat, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 0.98 | — | |
| Parallel DecodingModel=LLaMA-2-13B, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 0.97 | — | |
| Parallel DecodingModel=LLaMA-2-70B, Decoding Strategy=Greedy, Batch size=1, Shot configuration=1-shot, Maximum output length=1282026.02 | 0.96 | — |