Instructions to use chakkakrishna/falconfinetune with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use chakkakrishna/falconfinetune with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("ericzzz/falcon-rw-1b-chat") model = PeftModel.from_pretrained(base_model, "chakkakrishna/falconfinetune") - Notebooks
- Google Colab
- Kaggle
Download config.json from chakkakrishna/falconfinetune: direct link, hf CLI and curl.
- Browser
- Download file 752 Bytes
-
https://huggingface.co/chakkakrishna/falconfinetune/resolve/main/config.json
- Command line
-
hf download hf://chakkakrishna/falconfinetune/config.json
-
curl -L -o config.json https://huggingface.co/chakkakrishna/falconfinetune/resolve/main/config.json
752 Bytes
| { | |
| "_name_or_path": "ericzzz/falcon-rw-1b-chat", | |
| "alibi": true, | |
| "apply_residual_connection_post_layernorm": false, | |
| "architectures": [ | |
| "FalconForCausalLM" | |
| ], | |
| "attention_dropout": 0.0, | |
| "bias": true, | |
| "bos_token_id": 11, | |
| "eos_token_id": 11, | |
| "hidden_dropout": 0.0, | |
| "hidden_size": 2048, | |
| "initializer_range": 0.02, | |
| "layer_norm_epsilon": 1e-05, | |
| "max_position_embeddings": 2048, | |
| "model_type": "falcon", | |
| "multi_query": false, | |
| "new_decoder_architecture": false, | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 24, | |
| "num_kv_heads": 32, | |
| "parallel_attn": false, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.31.0", | |
| "use_cache": false, | |
| "vocab_size": 50304 | |
| } | |