# Language Modeling on Language Modeling Dataset PPL Llama-2-70B

Source: [Language Modeling on Language Modeling Dataset PPL Llama-2-70B](<https://www.sota2.com/research/sota/language-modeling-on-language-modeling-dataset-ppl-llama-2-70b>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-16T06:53:04.019Z

Evaluation of Language Modeling performance measured by Perplexity (PPL), specifically using the dataset context associated with the Llama-2-70B evaluation setup.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Language Modeling |
| Dataset (including any stated split) | Language Modeling Dataset PPL Llama-2-70B |
| Metric | Perplexity |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 7 of 7 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Perplexity](<https://www.sota2.com/research/sota/language-modeling-on-language-modeling-dataset-ppl-llama-2-70b.md?metric=Perplexity>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Perplexity | Result date | Report paper |
| --- | --- | --- | --- |
| Dense | 3.32 | Not recorded | 68fa79cc1e209131cea7f6b5 |
| TRSP-ℓ2 | 4.13 | Not recorded | 68fa79cc1e209131cea7f6b5 |
| TRSP-ℓ1 | 4.28 | Not recorded | 68fa79cc1e209131cea7f6b5 |
| ShortGPT | 4.85 | Not recorded | 68fa79cc1e209131cea7f6b5 |
| LaCo | 4.92 | Not recorded | 68fa79cc1e209131cea7f6b5 |
| Shortened LLaMA | 4.98 | Not recorded | 68fa79cc1e209131cea7f6b5 |
| SLEB | 5.06 | Not recorded | 68fa79cc1e209131cea7f6b5 |

### Dense

Perplexity: 3.32. Result ID: 69e0863a47f34ce924c938ae.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79cc1e209131cea7f6b5.md>)

Recorded evaluation configuration: {"Pruning Ratio":"0%"}. This record does not assert independent reproduction.

### TRSP-ℓ2

Perplexity: 4.13. Result ID: 69e0863a47f34ce924c938b3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79cc1e209131cea7f6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79cc1e209131cea7f6b5.md>)

Recorded evaluation configuration: {"Pruning Ratio":"25%"}. This record does not assert independent reproduction.

### TRSP-ℓ1

Perplexity: 4.28. Result ID: 69e0863a47f34ce924c938b4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79cc1e209131cea7f6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79cc1e209131cea7f6b5.md>)

Recorded evaluation configuration: {"Pruning Ratio":"25%"}. This record does not assert independent reproduction.

### ShortGPT

Perplexity: 4.85. Result ID: 69e0863a47f34ce924c938b0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79cc1e209131cea7f6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75391e209131cea668f7.md>)

Recorded evaluation configuration: {"Pruning Ratio":"25%"}. This record does not assert independent reproduction.

### LaCo

Perplexity: 4.92. Result ID: 69e0863a47f34ce924c938b1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79cc1e209131cea7f6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75321e209131cea66502.md>)

Recorded evaluation configuration: {"Pruning Ratio":"25%"}. This record does not assert independent reproduction.

### Shortened LLaMA

Perplexity: 4.98. Result ID: 69e0863a47f34ce924c938b2.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79cc1e209131cea7f6b5.md>)

Recorded evaluation configuration: {"Pruning Ratio":"25%"}. This record does not assert independent reproduction.

### SLEB

Perplexity: 5.06. Result ID: 69e0863a47f34ce924c938af.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79cc1e209131cea7f6b5.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76261e209131cea6b833.md>)

Recorded evaluation configuration: {"Pruning Ratio":"25%"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.