From 874d26b0b9b9bb2e2b761e1bedcd50b4eea14af9 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Quentin=20Gallou=C3=A9dec?= Date: Wed, 9 Apr 2025 01:06:09 +0000 Subject: [PATCH] tiny llama --- scripts/generate_tiny_models.py | 3 +++ 1 file changed, 3 insertions(+) diff --git a/scripts/generate_tiny_models.py b/scripts/generate_tiny_models.py index ea3b2df423..3c4a434001 100644 --- a/scripts/generate_tiny_models.py +++ b/scripts/generate_tiny_models.py @@ -41,6 +41,8 @@ GPTNeoXForCausalLM, Idefics2Config, Idefics2ForConditionalGeneration, + Llama4ForCausalLM, + Llama4TextConfig, LlamaConfig, LlamaForCausalLM, LlamaForSequenceClassification, @@ -114,6 +116,7 @@ def push_to_hub(model, tokenizer, prefix=None, suffix=None): ("meta-llama/Meta-Llama-3-8B-Instruct", LlamaConfig, LlamaForCausalLM, "3"), ("meta-llama/Llama-3.1-8B-Instruct", LlamaConfig, LlamaForCausalLM, "3.1"), ("meta-llama/Llama-3.2-1B-Instruct", LlamaConfig, LlamaForCausalLM, "3.2"), + ("meta-llama/Llama-4-Scout-17B-16E-Instruct", Llama4TextConfig, Llama4ForCausalLM, None), ("mistralai/Mistral-7B-Instruct-v0.1", MistralConfig, MistralForCausalLM, "0.1"), ("mistralai/Mistral-7B-Instruct-v0.2", MistralConfig, MistralForCausalLM, "0.2"), ("facebook/opt-1.3b", OPTConfig, OPTForCausalLM, None),