license, license_name, license_link, language, pipeline_tag, base_model
license license_name license_link language pipeline_tag base_model
other glm-4 https://huggingface.co/zai-org/glm-edge-1.5b-chat/raw/main/LICENSE
el
en
zh
text-generation
hereticness/heretic_glm-edge-1.5b-chat

GLM-Edge-1.5B-Chat-Greek

Quarduple fine-tuned version of this.

Improved Greek language skills, lack of coherence.

Extremely prone to hallunication and to surreal rhetoric.

1st finetune - CausalLM/GPT-4-Self-Instruct-Greek (1 epoch, 0,0002 learning rate)

2nd finetune - The first 10000 rows of enchatted/alpaca-cleaned-gpt4-el (1 epoch, 0,0002 learning rate)

3rd finetune - Rows 10500-23500 of enchatted/alpaca-cleaned-gpt4-el (1 epoch, 0,0001 learning rate)

4th finetune - aiuser3993/Llama-Krikri-8B-Instruct-Chat (1 epoch, 0,00005 learning rate)

License

GLM 4

S
Description
Model synced from source: aiuser3993/GLM-Edge-1.5B-Chat-Greek
Readme
1.3 MiB
0 Stars 1 Watchers 0 Forks
Languages
Jinja 100%