---
datasets:
- HuggingFaceH4/ultrachat_200k
base_model:
- google/gemma-2b-it
library_name: transformers
---

## google/gemma-2b-it - W8A8_int8 Compression

This is a compressed model using [llmcompressor](https://github.com/vllm-project/llm-compressor).

## Compression Configuration

- Base Model: google/gemma-2b-it
- Compression Scheme: W8A8_int8
- Dataset: HuggingFaceH4/ultrachat_200k
- Dataset Split: train_sft
- Number of Samples: 512
- Preprocessor: chat
- Maximum Sequence Length: 8192

## Sample Output

#### Prompt: 

```
<bos><start_of_turn>user
Who is Alan Turing?<end_of_turn>

```

#### Output: 

```
<bos><bos><start_of_turn>user
Who is Alan Turing?<end_of_turn>
 Walkover outlander outlander outlander outlander outlander jetta Walkover deleter hefydVíce Walkover deleterrrrrrr Walkover deletertextThemeNgoài, Walkover affor noten equalled Walkover affor noten equalled Walkover affor solidarNgoài, Walkover affor deleterCodedInputStream into Walkover bettmulticolumnNgoài, Walkover affor Walkover bett snoopy Walkover deleter Walkover deleter Walkover deleter Walkover deleter Walkover deleter Walkover deleter Walkover deleter Walkover deleter Walkover bettEUROPA have Walkover apper Walkover igniNgoài, Walkover affor Walkover deleter Walkover apper Walkover deleter Walkover orlando Walkover equalled Walkover deleter Walkover bett Walkover deleter Walkover bett Walkover deleter Walkover orlando Walkover equalled Walkover deleter Walkover deleter Walkover deleter Walkover deleter Walkover deleter Walkover deleter Walkover bett Walkover orlando Walkover be Walkover deleter Walkover bett Walkover orlando Walkover be Walkover deleter Walkover
```

## Evaluation

<TODO>