---
base_model:
- l3utterfly/tinyllama-1.1b-layla-v4
- appvoid/palmer-003
- sreeramajay/TinyLlama-1.1B-orca-v1.0
- Josephgflowers/TinyLlama-3T-Cinder-v1.3
- TinyLlama/TinyLlama-1.1B-intermediate-step-1195k-token-2.5T
- vihangd/DopeyTinyLlama-1.1B-v1
library_name: transformers
tags:
- mergekit
- merge

---
# merge

This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).

## Merge Details
### Merge Method

This model was merged using the [TIES](https://arxiv.org/abs/2306.01708) merge method using [TinyLlama/TinyLlama-1.1B-intermediate-step-1195k-token-2.5T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1195k-token-2.5T) as a base.

### Models Merged

The following models were included in the merge:
* [l3utterfly/tinyllama-1.1b-layla-v4](https://huggingface.co/l3utterfly/tinyllama-1.1b-layla-v4)
* [appvoid/palmer-003](https://huggingface.co/appvoid/palmer-003)
* [sreeramajay/TinyLlama-1.1B-orca-v1.0](https://huggingface.co/sreeramajay/TinyLlama-1.1B-orca-v1.0)
* [Josephgflowers/TinyLlama-3T-Cinder-v1.3](https://huggingface.co/Josephgflowers/TinyLlama-3T-Cinder-v1.3)
* [vihangd/DopeyTinyLlama-1.1B-v1](https://huggingface.co/vihangd/DopeyTinyLlama-1.1B-v1)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
models:
  - model: TinyLlama/TinyLlama-1.1B-intermediate-step-1195k-token-2.5T
    #no parameters necessary for base model
  - model: vihangd/DopeyTinyLlama-1.1B-v1
    parameters:
      density: 0.80
      weight: 0.30
  - model: l3utterfly/tinyllama-1.1b-layla-v4
    parameters:
      density: 0.25
      weight: 0.20
  - model: Josephgflowers/TinyLlama-3T-Cinder-v1.3
    parameters:
      density: 0.25
      weight: 0.10
  - model: sreeramajay/TinyLlama-1.1B-orca-v1.0
    parameters:
      density: 0.25
      weight: 0.10
  - model: appvoid/palmer-003
    parameters:
      density: 1
      weight: 0.30

merge_method: ties
base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-1195k-token-2.5T
parameters:
  normalize: true
  int8_mask: true
dtype: float16
```