From dd8bad06d0b4849a5a32dcb7f0f91b4a5be869f7 Mon Sep 17 00:00:00 2001 From: Wing Lian Date: Sat, 12 Apr 2025 07:25:11 -0700 Subject: [PATCH] remove strict=false from example yamls [skip ci] (#2523) [skip ci] --- examples/cerebras/btlm-ft.yml | 1 - examples/cerebras/qlora.yml | 1 - examples/code-llama/13b/lora.yml | 1 - examples/code-llama/13b/qlora.yml | 1 - examples/code-llama/34b/lora.yml | 1 - examples/code-llama/34b/qlora.yml | 1 - examples/code-llama/7b/lora.yml | 1 - examples/code-llama/7b/qlora.yml | 1 - examples/cohere/command-r-7b-qlora.yml | 1 - examples/dbrx/16bit-lora.yaml | 1 - examples/dbrx/8bit-lora.yaml | 1 - examples/dbrx/fft-ds-zero3.yaml | 1 - examples/deepseek-v2/fft-fsdp-16b.yaml | 1 - examples/deepseek-v2/qlora-fsdp-2_5.yaml | 1 - examples/falcon/config-7b-lora.yml | 1 - examples/falcon/config-7b-qlora.yml | 1 - examples/falcon/config-7b.yml | 1 - examples/gemma/qlora.yml | 1 - examples/gemma2/qlora.yml | 1 - examples/gemma2/reward-model.yaml | 1 - examples/gemma3/gemma-3-1b-qlora.yml | 1 - examples/gemma3/gemma-3-4b-qlora.yml | 1 - examples/gemma3/gemma-3-4b-vision-qlora.yml | 1 - examples/gptj/qlora.yml | 1 - examples/jamba/qlora.yaml | 1 - examples/jamba/qlora_deepspeed.yaml | 1 - examples/jamba/qlora_fsdp_large.yaml | 1 - examples/llama-2/fft_optimized.yml | 1 - examples/llama-2/gptq-lora.yml | 1 - examples/llama-2/lisa.yml | 1 - examples/llama-2/loftq.yml | 1 - examples/llama-2/lora.yml | 1 - examples/llama-2/qlora-fsdp.yml | 1 - examples/llama-2/qlora.yml | 1 - examples/llama-2/relora.yml | 1 - examples/llama-3-vision/lora-11b.yaml | 1 - examples/llama-3/fft-8b-liger-fsdp.yaml | 1 - examples/llama-3/fft-8b.yaml | 1 - examples/llama-3/instruct-dpo-lora-8b.yml | 1 - examples/llama-3/instruct-lora-8b.yml | 1 - examples/llama-3/lora-1b-deduplicate-dpo.yml | 1 - examples/llama-3/lora-1b-deduplicate-sft.yml | 1 - examples/llama-3/lora-1b-kernels.yml | 1 - examples/llama-3/lora-1b-ray.yml | 1 - examples/llama-3/lora-1b-sample-packing-sequentially.yml | 1 - examples/llama-3/lora-1b.yml | 1 - examples/llama-3/lora-8b.yml | 1 - examples/llama-3/qlora-1b-kto.yaml | 1 - examples/llama-3/qlora-1b.yml | 1 - examples/llama-3/qlora-fsdp-405b.yaml | 1 - examples/llama-3/qlora-fsdp-70b.yaml | 1 - examples/llama-3/qlora.yml | 1 - examples/llama-4/maverick-qlora-fsdp1.yaml | 1 - examples/llama-4/scout-qlora-fsdp1.yaml | 1 - examples/llama-4/scout-qlora-single-h100.yaml | 1 - examples/llama-4/scout-vision-qlora-fsdp.yaml | 1 - examples/llava/lora-7b.yaml | 1 - examples/mamba/config.yml | 1 - examples/mistral/bigstral-ds-zero3.yaml | 1 - examples/mistral/config.yml | 1 - examples/mistral/lora-mps.yml | 1 - examples/mistral/lora.yml | 1 - examples/mistral/mistral-dpo-qlora.yml | 1 - examples/mistral/mistral-qlora-fsdp.yml | 1 - examples/mistral/mistral-qlora-orpo.yml | 1 - examples/mistral/mistral-small-3.1-24B-lora.yml | 1 - examples/mistral/mixtral-8x22b-qlora-fsdp.yml | 1 - examples/mistral/mixtral-qlora-fsdp.yml | 1 - examples/mistral/mixtral.yml | 1 - examples/mistral/mixtral_22.yml | 1 - examples/mistral/qlora.yml | 1 - examples/openllama-3b/config.yml | 1 - examples/openllama-3b/lora.yml | 1 - examples/openllama-3b/qlora.yml | 1 - examples/phi/lora-3.5.yaml | 1 - examples/phi/phi-ft.yml | 1 - examples/phi/phi-qlora.yml | 1 - examples/phi/phi2-ft.yml | 1 - examples/phi/phi3-ft-fsdp.yml | 1 - examples/phi/phi3-ft.yml | 1 - examples/pixtral/lora-12b.yml | 1 - examples/qwen/lora.yml | 1 - examples/qwen/qlora.yml | 1 - examples/qwen/qwen2-moe-lora.yaml | 1 - examples/qwen/qwen2-moe-qlora.yaml | 1 - examples/qwen2-vl/lora-7b.yaml | 1 - examples/qwen2/dpo.yaml | 1 - examples/qwen2/prm.yaml | 1 - examples/qwen2/qlora-fsdp.yaml | 1 - examples/qwen2/reward-model.yaml | 1 - examples/stablelm-2/1.6b/fft.yml | 1 - examples/stablelm-2/1.6b/lora.yml | 1 - examples/starcoder2/qlora.yml | 1 - examples/tiny-llama/lora-mps.yml | 1 - examples/tiny-llama/lora.yml | 1 - examples/tiny-llama/pretrain.yml | 1 - examples/tiny-llama/qlora.yml | 1 - examples/xgen-7b/xgen-7b-8k-qlora.yml | 1 - examples/yi-34B-chat/qlora.yml | 1 - 99 files changed, 99 deletions(-) diff --git a/examples/cerebras/btlm-ft.yml b/examples/cerebras/btlm-ft.yml index 6190714b4..c9878779d 100644 --- a/examples/cerebras/btlm-ft.yml +++ b/examples/cerebras/btlm-ft.yml @@ -8,7 +8,6 @@ tokenizer_type: GPT2Tokenizer trust_remote_code: true tokenizer_use_fast: true tokenizer_legacy: true -strict: false push_dataset_to_hub: hf_use_auth_token: true datasets: diff --git a/examples/cerebras/qlora.yml b/examples/cerebras/qlora.yml index e74b2d675..55cc597f1 100644 --- a/examples/cerebras/qlora.yml +++ b/examples/cerebras/qlora.yml @@ -4,7 +4,6 @@ base_model: cerebras/Cerebras-GPT-1.3B load_in_8bit: false load_in_4bit: true -strict: false push_dataset_to_hub: datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/code-llama/13b/lora.yml b/examples/code-llama/13b/lora.yml index 6c205ae87..0ed2382ba 100644 --- a/examples/code-llama/13b/lora.yml +++ b/examples/code-llama/13b/lora.yml @@ -7,7 +7,6 @@ tokenizer_type: CodeLlamaTokenizer load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/code-llama/13b/qlora.yml b/examples/code-llama/13b/qlora.yml index 28f0275d3..22bd1691b 100644 --- a/examples/code-llama/13b/qlora.yml +++ b/examples/code-llama/13b/qlora.yml @@ -7,7 +7,6 @@ tokenizer_type: CodeLlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/code-llama/34b/lora.yml b/examples/code-llama/34b/lora.yml index 6024ce3f7..25dc9f421 100644 --- a/examples/code-llama/34b/lora.yml +++ b/examples/code-llama/34b/lora.yml @@ -7,7 +7,6 @@ tokenizer_type: CodeLlamaTokenizer load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/code-llama/34b/qlora.yml b/examples/code-llama/34b/qlora.yml index 56c276cc9..0e33e2a45 100644 --- a/examples/code-llama/34b/qlora.yml +++ b/examples/code-llama/34b/qlora.yml @@ -7,7 +7,6 @@ tokenizer_type: CodeLlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/code-llama/7b/lora.yml b/examples/code-llama/7b/lora.yml index 0eb20c244..d288b9f65 100644 --- a/examples/code-llama/7b/lora.yml +++ b/examples/code-llama/7b/lora.yml @@ -7,7 +7,6 @@ tokenizer_type: CodeLlamaTokenizer load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/code-llama/7b/qlora.yml b/examples/code-llama/7b/qlora.yml index f078f1398..de41c0123 100644 --- a/examples/code-llama/7b/qlora.yml +++ b/examples/code-llama/7b/qlora.yml @@ -7,7 +7,6 @@ tokenizer_type: CodeLlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/cohere/command-r-7b-qlora.yml b/examples/cohere/command-r-7b-qlora.yml index 8a2b6eacd..4a30e9a77 100644 --- a/examples/cohere/command-r-7b-qlora.yml +++ b/examples/cohere/command-r-7b-qlora.yml @@ -4,7 +4,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: false load_in_4bit: true -strict: false # huggingface repo chat_template: cohere diff --git a/examples/dbrx/16bit-lora.yaml b/examples/dbrx/16bit-lora.yaml index 1724c1426..852654d49 100644 --- a/examples/dbrx/16bit-lora.yaml +++ b/examples/dbrx/16bit-lora.yaml @@ -3,7 +3,6 @@ base_model: LnL-AI/dbrx-base-converted-v2 # hub_model_id: username/custom_model_name trust_remote_code: true -strict: false datasets: - path: tatsu-lab/alpaca diff --git a/examples/dbrx/8bit-lora.yaml b/examples/dbrx/8bit-lora.yaml index 308483adf..0b9402194 100644 --- a/examples/dbrx/8bit-lora.yaml +++ b/examples/dbrx/8bit-lora.yaml @@ -6,7 +6,6 @@ trust_remote_code: true load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: tatsu-lab/alpaca diff --git a/examples/dbrx/fft-ds-zero3.yaml b/examples/dbrx/fft-ds-zero3.yaml index 0fbb5b068..e42c16673 100644 --- a/examples/dbrx/fft-ds-zero3.yaml +++ b/examples/dbrx/fft-ds-zero3.yaml @@ -3,7 +3,6 @@ base_model: LnL-AI/dbrx-base-converted-v2 # hub_model_id: username/custom_model_name trust_remote_code: true -strict: false datasets: - path: tatsu-lab/alpaca diff --git a/examples/deepseek-v2/fft-fsdp-16b.yaml b/examples/deepseek-v2/fft-fsdp-16b.yaml index 3fe8691a3..0ed97db36 100644 --- a/examples/deepseek-v2/fft-fsdp-16b.yaml +++ b/examples/deepseek-v2/fft-fsdp-16b.yaml @@ -2,7 +2,6 @@ base_model: deepseek-ai/DeepSeek-V2-Lite # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name trust_remote_code: true -strict: false datasets: - path: tatsu-lab/alpaca diff --git a/examples/deepseek-v2/qlora-fsdp-2_5.yaml b/examples/deepseek-v2/qlora-fsdp-2_5.yaml index a554970b6..34dbeaafe 100644 --- a/examples/deepseek-v2/qlora-fsdp-2_5.yaml +++ b/examples/deepseek-v2/qlora-fsdp-2_5.yaml @@ -6,7 +6,6 @@ trust_remote_code: true load_in_8bit: false load_in_4bit: true -strict: false plugins: diff --git a/examples/falcon/config-7b-lora.yml b/examples/falcon/config-7b-lora.yml index 2d9240e8b..391d4dd94 100644 --- a/examples/falcon/config-7b-lora.yml +++ b/examples/falcon/config-7b-lora.yml @@ -11,7 +11,6 @@ trust_remote_code: true load_in_8bit: true load_in_4bit: false gptq: false -strict: false push_dataset_to_hub: datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/falcon/config-7b-qlora.yml b/examples/falcon/config-7b-qlora.yml index 78323db5f..a9af8574c 100644 --- a/examples/falcon/config-7b-qlora.yml +++ b/examples/falcon/config-7b-qlora.yml @@ -15,7 +15,6 @@ load_in_8bit: false # enable 4bit for QLoRA load_in_4bit: true gptq: false -strict: false push_dataset_to_hub: datasets: - path: QingyiSi/Alpaca-CoT diff --git a/examples/falcon/config-7b.yml b/examples/falcon/config-7b.yml index a796b89dd..3cc553daa 100644 --- a/examples/falcon/config-7b.yml +++ b/examples/falcon/config-7b.yml @@ -8,7 +8,6 @@ tokenizer_type: AutoTokenizer # required by falcon custom model code: https://huggingface.co/tiiuae/falcon-7b/tree/main trust_remote_code: true gptq: false -strict: false push_dataset_to_hub: datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/gemma/qlora.yml b/examples/gemma/qlora.yml index 505564269..2738112b4 100644 --- a/examples/gemma/qlora.yml +++ b/examples/gemma/qlora.yml @@ -8,7 +8,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: false load_in_4bit: true -strict: false # huggingface repo datasets: diff --git a/examples/gemma2/qlora.yml b/examples/gemma2/qlora.yml index afba83552..cb96a32c1 100644 --- a/examples/gemma2/qlora.yml +++ b/examples/gemma2/qlora.yml @@ -7,7 +7,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: false load_in_4bit: true -strict: false # huggingface repo chat_template: gemma diff --git a/examples/gemma2/reward-model.yaml b/examples/gemma2/reward-model.yaml index d828af939..ce01a4572 100644 --- a/examples/gemma2/reward-model.yaml +++ b/examples/gemma2/reward-model.yaml @@ -5,7 +5,6 @@ num_labels: 1 tokenizer_type: AutoTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false reward_model: true chat_template: gemma diff --git a/examples/gemma3/gemma-3-1b-qlora.yml b/examples/gemma3/gemma-3-1b-qlora.yml index 732b914a8..44310558c 100644 --- a/examples/gemma3/gemma-3-1b-qlora.yml +++ b/examples/gemma3/gemma-3-1b-qlora.yml @@ -10,7 +10,6 @@ ddp_find_unused_parameters: true load_in_8bit: false load_in_4bit: true -strict: false # huggingface repo chat_template: gemma3 diff --git a/examples/gemma3/gemma-3-4b-qlora.yml b/examples/gemma3/gemma-3-4b-qlora.yml index 85e5dce68..29f8cc1e1 100644 --- a/examples/gemma3/gemma-3-4b-qlora.yml +++ b/examples/gemma3/gemma-3-4b-qlora.yml @@ -1,5 +1,4 @@ base_model: google/gemma-3-4b-it -strict: false load_in_4bit: true diff --git a/examples/gemma3/gemma-3-4b-vision-qlora.yml b/examples/gemma3/gemma-3-4b-vision-qlora.yml index 92273380c..3fd9eb5f0 100644 --- a/examples/gemma3/gemma-3-4b-vision-qlora.yml +++ b/examples/gemma3/gemma-3-4b-vision-qlora.yml @@ -1,6 +1,5 @@ base_model: google/gemma-3-4b-it processor_type: AutoProcessor -strict: false load_in_4bit: true diff --git a/examples/gptj/qlora.yml b/examples/gptj/qlora.yml index 086d425b5..c3cf9f973 100644 --- a/examples/gptj/qlora.yml +++ b/examples/gptj/qlora.yml @@ -4,7 +4,6 @@ base_model: EleutherAI/gpt-j-6b load_in_8bit: false load_in_4bit: true -strict: false push_dataset_to_hub: datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/jamba/qlora.yaml b/examples/jamba/qlora.yaml index 7d642cb0a..2cb0eea41 100644 --- a/examples/jamba/qlora.yaml +++ b/examples/jamba/qlora.yaml @@ -6,7 +6,6 @@ trust_remote_code: true load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/jamba/qlora_deepspeed.yaml b/examples/jamba/qlora_deepspeed.yaml index d983dc391..d13ce6483 100644 --- a/examples/jamba/qlora_deepspeed.yaml +++ b/examples/jamba/qlora_deepspeed.yaml @@ -5,7 +5,6 @@ trust_remote_code: true load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/jamba/qlora_fsdp_large.yaml b/examples/jamba/qlora_fsdp_large.yaml index a968d99d7..6badaba19 100644 --- a/examples/jamba/qlora_fsdp_large.yaml +++ b/examples/jamba/qlora_fsdp_large.yaml @@ -5,7 +5,6 @@ tokenizer_type: AutoTokenizer # hub_model_id: username/custom_model_name load_in_4bit: true -strict: false use_tensorboard: true chat_template: jamba datasets: diff --git a/examples/llama-2/fft_optimized.yml b/examples/llama-2/fft_optimized.yml index fd78fbfee..86b1b6a21 100644 --- a/examples/llama-2/fft_optimized.yml +++ b/examples/llama-2/fft_optimized.yml @@ -4,7 +4,6 @@ model_type: LlamaForCausalLM tokenizer_type: LlamaTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/llama-2/gptq-lora.yml b/examples/llama-2/gptq-lora.yml index ad2dbd9cf..0f1b34016 100644 --- a/examples/llama-2/gptq-lora.yml +++ b/examples/llama-2/gptq-lora.yml @@ -10,7 +10,6 @@ gptq_disable_exllama: true tokenizer_use_fast: true tokenizer_legacy: true -strict: false push_dataset_to_hub: hf_use_auth_token: true datasets: diff --git a/examples/llama-2/lisa.yml b/examples/llama-2/lisa.yml index 585fa9428..a76a792ae 100644 --- a/examples/llama-2/lisa.yml +++ b/examples/llama-2/lisa.yml @@ -4,7 +4,6 @@ model_type: LlamaForCausalLM tokenizer_type: LlamaTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/llama-2/loftq.yml b/examples/llama-2/loftq.yml index bf32c7b27..22dbf2d99 100644 --- a/examples/llama-2/loftq.yml +++ b/examples/llama-2/loftq.yml @@ -4,7 +4,6 @@ model_type: LlamaForCausalLM tokenizer_type: LlamaTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/llama-2/lora.yml b/examples/llama-2/lora.yml index 3ef607ab4..679aed3a9 100644 --- a/examples/llama-2/lora.yml +++ b/examples/llama-2/lora.yml @@ -7,7 +7,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/llama-2/qlora-fsdp.yml b/examples/llama-2/qlora-fsdp.yml index 759f08024..a42eabd4b 100644 --- a/examples/llama-2/qlora-fsdp.yml +++ b/examples/llama-2/qlora-fsdp.yml @@ -7,7 +7,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: yahma/alpaca-cleaned diff --git a/examples/llama-2/qlora.yml b/examples/llama-2/qlora.yml index c678a0042..de65928bc 100644 --- a/examples/llama-2/qlora.yml +++ b/examples/llama-2/qlora.yml @@ -7,7 +7,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/llama-2/relora.yml b/examples/llama-2/relora.yml index 6c943d009..e0a5f7068 100644 --- a/examples/llama-2/relora.yml +++ b/examples/llama-2/relora.yml @@ -5,7 +5,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/llama-3-vision/lora-11b.yaml b/examples/llama-3-vision/lora-11b.yaml index 4431878fa..f4883e903 100644 --- a/examples/llama-3-vision/lora-11b.yaml +++ b/examples/llama-3-vision/lora-11b.yaml @@ -4,7 +4,6 @@ processor_type: AutoProcessor # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false # these 3 lines are needed for now to handle vision chat templates w images skip_prepare_dataset: true diff --git a/examples/llama-3/fft-8b-liger-fsdp.yaml b/examples/llama-3/fft-8b-liger-fsdp.yaml index 50169879c..eccfa6d8c 100644 --- a/examples/llama-3/fft-8b-liger-fsdp.yaml +++ b/examples/llama-3/fft-8b-liger-fsdp.yaml @@ -9,7 +9,6 @@ liger_rms_norm: true liger_glu_activation: true liger_fused_linear_cross_entropy: true -strict: false chat_template: llama3 datasets: diff --git a/examples/llama-3/fft-8b.yaml b/examples/llama-3/fft-8b.yaml index 4452a6e3d..fdae3e6c4 100644 --- a/examples/llama-3/fft-8b.yaml +++ b/examples/llama-3/fft-8b.yaml @@ -1,7 +1,6 @@ base_model: NousResearch/Meta-Llama-3.1-8B # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: tatsu-lab/alpaca diff --git a/examples/llama-3/instruct-dpo-lora-8b.yml b/examples/llama-3/instruct-dpo-lora-8b.yml index a1b923fb6..13082294f 100644 --- a/examples/llama-3/instruct-dpo-lora-8b.yml +++ b/examples/llama-3/instruct-dpo-lora-8b.yml @@ -7,7 +7,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: true load_in_4bit: false -strict: false chat_template: llama3 rl: dpo diff --git a/examples/llama-3/instruct-lora-8b.yml b/examples/llama-3/instruct-lora-8b.yml index 362bda9aa..acab862f6 100644 --- a/examples/llama-3/instruct-lora-8b.yml +++ b/examples/llama-3/instruct-lora-8b.yml @@ -7,7 +7,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: true load_in_4bit: false -strict: false chat_template: llama3 datasets: diff --git a/examples/llama-3/lora-1b-deduplicate-dpo.yml b/examples/llama-3/lora-1b-deduplicate-dpo.yml index e4b2a5244..10e9747cb 100644 --- a/examples/llama-3/lora-1b-deduplicate-dpo.yml +++ b/examples/llama-3/lora-1b-deduplicate-dpo.yml @@ -7,7 +7,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: true load_in_4bit: false -strict: false chat_template: llama3 rl: dpo diff --git a/examples/llama-3/lora-1b-deduplicate-sft.yml b/examples/llama-3/lora-1b-deduplicate-sft.yml index b8c21fafb..630ec92f6 100644 --- a/examples/llama-3/lora-1b-deduplicate-sft.yml +++ b/examples/llama-3/lora-1b-deduplicate-sft.yml @@ -7,7 +7,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/llama-3/lora-1b-kernels.yml b/examples/llama-3/lora-1b-kernels.yml index b76f03801..a2d07ca49 100644 --- a/examples/llama-3/lora-1b-kernels.yml +++ b/examples/llama-3/lora-1b-kernels.yml @@ -1,7 +1,6 @@ base_model: NousResearch/Llama-3.2-1B # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/llama-3/lora-1b-ray.yml b/examples/llama-3/lora-1b-ray.yml index 199fe3b5d..bb23164eb 100644 --- a/examples/llama-3/lora-1b-ray.yml +++ b/examples/llama-3/lora-1b-ray.yml @@ -1,7 +1,6 @@ base_model: NousResearch/Llama-3.2-1B # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/llama-3/lora-1b-sample-packing-sequentially.yml b/examples/llama-3/lora-1b-sample-packing-sequentially.yml index a027673ab..769dd32e6 100644 --- a/examples/llama-3/lora-1b-sample-packing-sequentially.yml +++ b/examples/llama-3/lora-1b-sample-packing-sequentially.yml @@ -7,7 +7,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/llama-3/lora-1b.yml b/examples/llama-3/lora-1b.yml index 8a536260a..c31a9f39a 100644 --- a/examples/llama-3/lora-1b.yml +++ b/examples/llama-3/lora-1b.yml @@ -1,7 +1,6 @@ base_model: NousResearch/Llama-3.2-1B # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/llama-3/lora-8b.yml b/examples/llama-3/lora-8b.yml index 700dd4614..ad50cd38a 100644 --- a/examples/llama-3/lora-8b.yml +++ b/examples/llama-3/lora-8b.yml @@ -7,7 +7,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/llama-3/qlora-1b-kto.yaml b/examples/llama-3/qlora-1b-kto.yaml index 0dc37b40a..89a51ea68 100644 --- a/examples/llama-3/qlora-1b-kto.yaml +++ b/examples/llama-3/qlora-1b-kto.yaml @@ -4,7 +4,6 @@ base_model: meta-llama/Llama-3.2-1B load_in_8bit: false load_in_4bit: true -strict: false rl: kto rl_beta: 0.5 diff --git a/examples/llama-3/qlora-1b.yml b/examples/llama-3/qlora-1b.yml index c42dd2238..5c8fe6628 100644 --- a/examples/llama-3/qlora-1b.yml +++ b/examples/llama-3/qlora-1b.yml @@ -4,7 +4,6 @@ base_model: NousResearch/Llama-3.2-1B load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/llama-3/qlora-fsdp-405b.yaml b/examples/llama-3/qlora-fsdp-405b.yaml index 75c8f5973..2b7d51925 100644 --- a/examples/llama-3/qlora-fsdp-405b.yaml +++ b/examples/llama-3/qlora-fsdp-405b.yaml @@ -5,7 +5,6 @@ tokenizer_type: AutoTokenizer # hub_model_id: username/custom_model_name load_in_4bit: true -strict: false datasets: - path: tatsu-lab/alpaca diff --git a/examples/llama-3/qlora-fsdp-70b.yaml b/examples/llama-3/qlora-fsdp-70b.yaml index c4889d643..412b6721c 100644 --- a/examples/llama-3/qlora-fsdp-70b.yaml +++ b/examples/llama-3/qlora-fsdp-70b.yaml @@ -7,7 +7,6 @@ tokenizer_type: AutoTokenizer # PreTrainedTokenizerFast load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: tatsu-lab/alpaca diff --git a/examples/llama-3/qlora.yml b/examples/llama-3/qlora.yml index 607deb896..4cc9fc3db 100644 --- a/examples/llama-3/qlora.yml +++ b/examples/llama-3/qlora.yml @@ -7,7 +7,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: aaditya/alpaca_subset_1 diff --git a/examples/llama-4/maverick-qlora-fsdp1.yaml b/examples/llama-4/maverick-qlora-fsdp1.yaml index 232afc73e..2be94f4ef 100644 --- a/examples/llama-4/maverick-qlora-fsdp1.yaml +++ b/examples/llama-4/maverick-qlora-fsdp1.yaml @@ -3,7 +3,6 @@ model_type: Llama4ForConditionalGeneration # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false plugins: - axolotl.integrations.liger.LigerPlugin diff --git a/examples/llama-4/scout-qlora-fsdp1.yaml b/examples/llama-4/scout-qlora-fsdp1.yaml index ad2e46786..eeae872a6 100644 --- a/examples/llama-4/scout-qlora-fsdp1.yaml +++ b/examples/llama-4/scout-qlora-fsdp1.yaml @@ -3,7 +3,6 @@ model_type: Llama4ForConditionalGeneration # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false # torch_compile: true plugins: diff --git a/examples/llama-4/scout-qlora-single-h100.yaml b/examples/llama-4/scout-qlora-single-h100.yaml index 23a3a2195..17ad70634 100644 --- a/examples/llama-4/scout-qlora-single-h100.yaml +++ b/examples/llama-4/scout-qlora-single-h100.yaml @@ -3,7 +3,6 @@ model_type: Llama4ForConditionalGeneration # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false plugins: - axolotl.integrations.liger.LigerPlugin diff --git a/examples/llama-4/scout-vision-qlora-fsdp.yaml b/examples/llama-4/scout-vision-qlora-fsdp.yaml index 8b8c9abd1..eff708e4d 100644 --- a/examples/llama-4/scout-vision-qlora-fsdp.yaml +++ b/examples/llama-4/scout-vision-qlora-fsdp.yaml @@ -4,7 +4,6 @@ processor_type: Llama4Processor # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false # these 3 lines are needed for now to handle vision chat templates w images skip_prepare_dataset: true diff --git a/examples/llava/lora-7b.yaml b/examples/llava/lora-7b.yaml index 68e463585..54edd04dc 100644 --- a/examples/llava/lora-7b.yaml +++ b/examples/llava/lora-7b.yaml @@ -1,6 +1,5 @@ base_model: llava-hf/llava-1.5-7b-hf processor_type: AutoProcessor -strict: false # these 3 lines are needed for now to handle vision chat templates w images skip_prepare_dataset: true diff --git a/examples/mamba/config.yml b/examples/mamba/config.yml index 3a114bac7..3d4583932 100644 --- a/examples/mamba/config.yml +++ b/examples/mamba/config.yml @@ -5,7 +5,6 @@ tokenizer_type: AutoTokenizer tokenizer_config: EleutherAI/gpt-neox-20b # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/mistral/bigstral-ds-zero3.yaml b/examples/mistral/bigstral-ds-zero3.yaml index bef989932..f626a92a1 100644 --- a/examples/mistral/bigstral-ds-zero3.yaml +++ b/examples/mistral/bigstral-ds-zero3.yaml @@ -6,7 +6,6 @@ tokenizer_type: LlamaTokenizer # hub_model_id: username/custom_model_name trust_remote_code: true -strict: false unfrozen_parameters: - ^lm_head.weight$ diff --git a/examples/mistral/config.yml b/examples/mistral/config.yml index c58e2c954..15edffb44 100644 --- a/examples/mistral/config.yml +++ b/examples/mistral/config.yml @@ -4,7 +4,6 @@ model_type: MistralForCausalLM tokenizer_type: LlamaTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/mistral/lora-mps.yml b/examples/mistral/lora-mps.yml index ba61cac11..e6f46affb 100644 --- a/examples/mistral/lora-mps.yml +++ b/examples/mistral/lora-mps.yml @@ -4,7 +4,6 @@ model_type: MistralForCausalLM tokenizer_type: LlamaTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/mistral/lora.yml b/examples/mistral/lora.yml index b55c66715..9af4274fd 100644 --- a/examples/mistral/lora.yml +++ b/examples/mistral/lora.yml @@ -7,7 +7,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/mistral/mistral-dpo-qlora.yml b/examples/mistral/mistral-dpo-qlora.yml index 6446cacb8..af707973f 100644 --- a/examples/mistral/mistral-dpo-qlora.yml +++ b/examples/mistral/mistral-dpo-qlora.yml @@ -12,7 +12,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false chat_template: chatml rl: dpo diff --git a/examples/mistral/mistral-qlora-fsdp.yml b/examples/mistral/mistral-qlora-fsdp.yml index 5825ac749..e234b19a2 100644 --- a/examples/mistral/mistral-qlora-fsdp.yml +++ b/examples/mistral/mistral-qlora-fsdp.yml @@ -9,7 +9,6 @@ trust_remote_code: true load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: tatsu-lab/alpaca diff --git a/examples/mistral/mistral-qlora-orpo.yml b/examples/mistral/mistral-qlora-orpo.yml index 9c6ae74ef..6c0212b7c 100644 --- a/examples/mistral/mistral-qlora-orpo.yml +++ b/examples/mistral/mistral-qlora-orpo.yml @@ -7,7 +7,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false rl: orpo orpo_alpha: 0.1 diff --git a/examples/mistral/mistral-small-3.1-24B-lora.yml b/examples/mistral/mistral-small-3.1-24B-lora.yml index 0e6b4402d..198b3f373 100644 --- a/examples/mistral/mistral-small-3.1-24B-lora.yml +++ b/examples/mistral/mistral-small-3.1-24B-lora.yml @@ -1,6 +1,5 @@ base_model: mistralai/Mistral-Small-3.1-24B-Instruct-2503 processor_type: AutoProcessor -strict: false load_in_8bit: true diff --git a/examples/mistral/mixtral-8x22b-qlora-fsdp.yml b/examples/mistral/mixtral-8x22b-qlora-fsdp.yml index e29b6392a..af6ba5a76 100644 --- a/examples/mistral/mixtral-8x22b-qlora-fsdp.yml +++ b/examples/mistral/mixtral-8x22b-qlora-fsdp.yml @@ -7,7 +7,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: tatsu-lab/alpaca diff --git a/examples/mistral/mixtral-qlora-fsdp.yml b/examples/mistral/mixtral-qlora-fsdp.yml index 40bb5d5d1..b1843a138 100644 --- a/examples/mistral/mixtral-qlora-fsdp.yml +++ b/examples/mistral/mixtral-qlora-fsdp.yml @@ -9,7 +9,6 @@ trust_remote_code: true load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: tatsu-lab/alpaca diff --git a/examples/mistral/mixtral.yml b/examples/mistral/mixtral.yml index eefd2456d..4c256420c 100644 --- a/examples/mistral/mixtral.yml +++ b/examples/mistral/mixtral.yml @@ -9,7 +9,6 @@ trust_remote_code: true load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: tatsu-lab/alpaca diff --git a/examples/mistral/mixtral_22.yml b/examples/mistral/mixtral_22.yml index 1debd793a..25e1d7155 100644 --- a/examples/mistral/mixtral_22.yml +++ b/examples/mistral/mixtral_22.yml @@ -6,7 +6,6 @@ tokenizer_type: LlamaTokenizer # hub_model_id: username/custom_model_name trust_remote_code: true -strict: false unfrozen_parameters: - ^lm_head.weight$ diff --git a/examples/mistral/qlora.yml b/examples/mistral/qlora.yml index 921f3fe7b..607e33701 100644 --- a/examples/mistral/qlora.yml +++ b/examples/mistral/qlora.yml @@ -7,7 +7,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/openllama-3b/config.yml b/examples/openllama-3b/config.yml index 7a1e2d9a5..17eeb73ae 100644 --- a/examples/openllama-3b/config.yml +++ b/examples/openllama-3b/config.yml @@ -4,7 +4,6 @@ model_type: LlamaForCausalLM tokenizer_type: LlamaTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false push_dataset_to_hub: datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/openllama-3b/lora.yml b/examples/openllama-3b/lora.yml index c1c597b73..073117f11 100644 --- a/examples/openllama-3b/lora.yml +++ b/examples/openllama-3b/lora.yml @@ -7,7 +7,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: true load_in_4bit: false -strict: false push_dataset_to_hub: datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/openllama-3b/qlora.yml b/examples/openllama-3b/qlora.yml index e9c71efd1..b4fca2c07 100644 --- a/examples/openllama-3b/qlora.yml +++ b/examples/openllama-3b/qlora.yml @@ -7,7 +7,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false push_dataset_to_hub: datasets: - path: teknium/GPT4-LLM-Cleaned diff --git a/examples/phi/lora-3.5.yaml b/examples/phi/lora-3.5.yaml index 2ecb9d28d..ad4ce9cd4 100644 --- a/examples/phi/lora-3.5.yaml +++ b/examples/phi/lora-3.5.yaml @@ -7,7 +7,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: true load_in_4bit: false -strict: false chat_template: phi_3 datasets: diff --git a/examples/phi/phi-ft.yml b/examples/phi/phi-ft.yml index 886671a60..1562a7353 100644 --- a/examples/phi/phi-ft.yml +++ b/examples/phi/phi-ft.yml @@ -4,7 +4,6 @@ model_type: AutoModelForCausalLM tokenizer_type: AutoTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: garage-bAInd/Open-Platypus diff --git a/examples/phi/phi-qlora.yml b/examples/phi/phi-qlora.yml index a1cbf8a52..4cd53db97 100644 --- a/examples/phi/phi-qlora.yml +++ b/examples/phi/phi-qlora.yml @@ -7,7 +7,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: garage-bAInd/Open-Platypus diff --git a/examples/phi/phi2-ft.yml b/examples/phi/phi2-ft.yml index 480017a39..ca733cc71 100644 --- a/examples/phi/phi2-ft.yml +++ b/examples/phi/phi2-ft.yml @@ -4,7 +4,6 @@ model_type: AutoModelForCausalLM tokenizer_type: AutoTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: garage-bAInd/Open-Platypus diff --git a/examples/phi/phi3-ft-fsdp.yml b/examples/phi/phi3-ft-fsdp.yml index 766db76f6..d0d14fea6 100644 --- a/examples/phi/phi3-ft-fsdp.yml +++ b/examples/phi/phi3-ft-fsdp.yml @@ -4,7 +4,6 @@ model_type: AutoModelForCausalLM tokenizer_type: AutoTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/phi/phi3-ft.yml b/examples/phi/phi3-ft.yml index 0a5da2e23..17c48da6f 100644 --- a/examples/phi/phi3-ft.yml +++ b/examples/phi/phi3-ft.yml @@ -7,7 +7,6 @@ tokenizer_type: AutoTokenizer # hub_model_id: username/custom_model_name chat_template: phi_3 -strict: false datasets: - path: garage-bAInd/Open-Platypus diff --git a/examples/pixtral/lora-12b.yml b/examples/pixtral/lora-12b.yml index d3b3efd70..dec8e4b5e 100644 --- a/examples/pixtral/lora-12b.yml +++ b/examples/pixtral/lora-12b.yml @@ -1,6 +1,5 @@ base_model: mistral-community/pixtral-12b processor_type: AutoProcessor -strict: false # these 3 lines are needed for now to handle vision chat templates w images skip_prepare_dataset: true diff --git a/examples/qwen/lora.yml b/examples/qwen/lora.yml index 23385d236..9a2843236 100644 --- a/examples/qwen/lora.yml +++ b/examples/qwen/lora.yml @@ -9,7 +9,6 @@ trust_remote_code: true load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/qwen/qlora.yml b/examples/qwen/qlora.yml index 854a682fe..5f85b44dd 100644 --- a/examples/qwen/qlora.yml +++ b/examples/qwen/qlora.yml @@ -9,7 +9,6 @@ trust_remote_code: true load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/qwen/qwen2-moe-lora.yaml b/examples/qwen/qwen2-moe-lora.yaml index a2a1e4d25..afce443a0 100644 --- a/examples/qwen/qwen2-moe-lora.yaml +++ b/examples/qwen/qwen2-moe-lora.yaml @@ -3,7 +3,6 @@ base_model: Qwen/Qwen1.5-MoE-A2.7B # hub_model_id: username/custom_model_name trust_remote_code: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/qwen/qwen2-moe-qlora.yaml b/examples/qwen/qwen2-moe-qlora.yaml index b1ab131a6..92a6842cf 100644 --- a/examples/qwen/qwen2-moe-qlora.yaml +++ b/examples/qwen/qwen2-moe-qlora.yaml @@ -6,7 +6,6 @@ trust_remote_code: true load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/qwen2-vl/lora-7b.yaml b/examples/qwen2-vl/lora-7b.yaml index 3d0b10adf..55773bc3d 100644 --- a/examples/qwen2-vl/lora-7b.yaml +++ b/examples/qwen2-vl/lora-7b.yaml @@ -1,6 +1,5 @@ base_model: Qwen/Qwen2-VL-7B-Instruct processor_type: AutoProcessor -strict: false # these 3 lines are needed for now to handle vision chat templates w images skip_prepare_dataset: true diff --git a/examples/qwen2/dpo.yaml b/examples/qwen2/dpo.yaml index 8df7f9dc4..3547c6c98 100644 --- a/examples/qwen2/dpo.yaml +++ b/examples/qwen2/dpo.yaml @@ -2,7 +2,6 @@ base_model: Qwen/Qwen2.5-0.5B # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false chat_template: qwen_25 rl: dpo diff --git a/examples/qwen2/prm.yaml b/examples/qwen2/prm.yaml index 669f8e2db..4afa24f3c 100644 --- a/examples/qwen2/prm.yaml +++ b/examples/qwen2/prm.yaml @@ -5,7 +5,6 @@ num_labels: 2 tokenizer_type: AutoTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false process_reward_model: true chat_template: diff --git a/examples/qwen2/qlora-fsdp.yaml b/examples/qwen2/qlora-fsdp.yaml index 1f2ed83b1..ed2670ab6 100644 --- a/examples/qwen2/qlora-fsdp.yaml +++ b/examples/qwen2/qlora-fsdp.yaml @@ -6,7 +6,6 @@ trust_remote_code: true load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: tatsu-lab/alpaca diff --git a/examples/qwen2/reward-model.yaml b/examples/qwen2/reward-model.yaml index fcbb9867e..822407a1f 100644 --- a/examples/qwen2/reward-model.yaml +++ b/examples/qwen2/reward-model.yaml @@ -5,7 +5,6 @@ num_labels: 1 tokenizer_type: AutoTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false reward_model: true chat_template: qwen_25 diff --git a/examples/stablelm-2/1.6b/fft.yml b/examples/stablelm-2/1.6b/fft.yml index f26714856..9b45b399f 100644 --- a/examples/stablelm-2/1.6b/fft.yml +++ b/examples/stablelm-2/1.6b/fft.yml @@ -6,7 +6,6 @@ tokenizer_type: AutoTokenizer # hub_model_id: username/custom_model_name trust_remote_code: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/stablelm-2/1.6b/lora.yml b/examples/stablelm-2/1.6b/lora.yml index aaa9908d1..31e5ad933 100644 --- a/examples/stablelm-2/1.6b/lora.yml +++ b/examples/stablelm-2/1.6b/lora.yml @@ -9,7 +9,6 @@ trust_remote_code: true load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/starcoder2/qlora.yml b/examples/starcoder2/qlora.yml index 8b21d7145..18d85f9c3 100644 --- a/examples/starcoder2/qlora.yml +++ b/examples/starcoder2/qlora.yml @@ -4,7 +4,6 @@ base_model: bigcode/starcoder2-3b load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/tiny-llama/lora-mps.yml b/examples/tiny-llama/lora-mps.yml index 8654a39bb..66cf7cfb3 100644 --- a/examples/tiny-llama/lora-mps.yml +++ b/examples/tiny-llama/lora-mps.yml @@ -7,7 +7,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/tiny-llama/lora.yml b/examples/tiny-llama/lora.yml index 64b2360d3..90998880f 100644 --- a/examples/tiny-llama/lora.yml +++ b/examples/tiny-llama/lora.yml @@ -6,7 +6,6 @@ tokenizer_type: AutoTokenizer load_in_8bit: true load_in_4bit: false -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/tiny-llama/pretrain.yml b/examples/tiny-llama/pretrain.yml index 2984c52ae..5b3706bcb 100644 --- a/examples/tiny-llama/pretrain.yml +++ b/examples/tiny-llama/pretrain.yml @@ -4,7 +4,6 @@ model_type: LlamaForCausalLM tokenizer_type: LlamaTokenizer # Automatically upload checkpoint and final model to HF # hub_model_id: username/custom_model_name -strict: false max_steps: 200 pretraining_dataset: diff --git a/examples/tiny-llama/qlora.yml b/examples/tiny-llama/qlora.yml index 79e3164a5..8b2a4565a 100644 --- a/examples/tiny-llama/qlora.yml +++ b/examples/tiny-llama/qlora.yml @@ -7,7 +7,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false datasets: - path: mhenrichsen/alpaca_2k_test diff --git a/examples/xgen-7b/xgen-7b-8k-qlora.yml b/examples/xgen-7b/xgen-7b-8k-qlora.yml index f4ff589e0..48066b130 100644 --- a/examples/xgen-7b/xgen-7b-8k-qlora.yml +++ b/examples/xgen-7b/xgen-7b-8k-qlora.yml @@ -13,7 +13,6 @@ load_in_8bit: false # enable 4bit for QLoRA load_in_4bit: true gptq: false -strict: false push_dataset_to_hub: datasets: - path: timdettmers/openassistant-guanaco diff --git a/examples/yi-34B-chat/qlora.yml b/examples/yi-34B-chat/qlora.yml index de79ed6ce..a0a95d86f 100644 --- a/examples/yi-34B-chat/qlora.yml +++ b/examples/yi-34B-chat/qlora.yml @@ -7,7 +7,6 @@ tokenizer_type: LlamaTokenizer load_in_8bit: false load_in_4bit: true -strict: false sequence_len: 1024 bf16: auto tf32: false