commit cd0daaed0870197c09f480e676e48c1e403de92e Author: ModelHub XC Date: Sat May 2 22:17:54 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: double7/vicuna-160m Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..a6344aa --- /dev/null +++ b/.gitattributes @@ -0,0 +1,35 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000..beaed7d --- /dev/null +++ b/README.md @@ -0,0 +1,22 @@ +--- +license: apache-2.0 +datasets: +- anon8231489123/ShareGPT_Vicuna_unfiltered +language: +- en +pipeline_tag: text-generation +--- +## Model description +This is a Vicuna-like model with only 160M parameters, which is fine-tuned from [LLaMA-160m](https://huggingface.co/JackFram/llama-160m) on [ShareGPT](https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered) data. + +The training setup follows the [Vicuna suite](https://github.com/lm-sys/FastChat). + +The model is mainly developed as a base Small Speculative Model in [MCSD paper](https://arxiv.org/pdf/2401.06706.pdf). As a comparison, it can be better aligned to the Vicuna models than LLaMA-160m with little loss of alignment to the LLaMA models. + +| Draft Model | Target Model | Alignment | +| -------------- | ------------- | --------- | +| LLaMA-68/160M | LLaMA-13/33B | 😃 | +| LLaMA-68/160M | Vicuna-13/33B | 😟 | +| Vicuna-68/160M | LLaMA-13/33B | 😃 | +| Vicuna-68/160M | Vicuna-13/33B | 😃 | + diff --git a/config.json b/config.json new file mode 100644 index 0000000..f5c3b6d --- /dev/null +++ b/config.json @@ -0,0 +1,29 @@ +{ + "_name_or_path": "/data0/yangs/llama-160m", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 0, + "eos_token_id": 2, + "hidden_act": "silu", + "hidden_size": 768, + "initializer_range": 0.02, + "intermediate_size": 3072, + "max_position_embeddings": 2048, + "model_type": "llama", + "num_attention_heads": 12, + "num_hidden_layers": 12, + "num_key_value_heads": 12, + "pad_token_id": 1, + "pretraining_tp": 1, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 10000.0, + "tie_word_embeddings": false, + "torch_dtype": "float32", + "transformers_version": "4.37.0.dev0", + "use_cache": true, + "vocab_size": 32000 +} diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000..d6af2fe --- /dev/null +++ b/generation_config.json @@ -0,0 +1,7 @@ +{ + "_from_model_config": true, + "bos_token_id": 0, + "eos_token_id": 2, + "pad_token_id": 1, + "transformers_version": "4.37.0.dev0" +} diff --git a/model.safetensors b/model.safetensors new file mode 100644 index 0000000..e96d994 --- /dev/null +++ b/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2c8df33c29d8cf3227c9953abf50f37f91bd8516a73ab3dd0a3030ad76fe4f4d +size 649681952 diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000..f928b24 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,24 @@ +{ + "bos_token": { + "content": "", + "lstrip": false, + "normalized": true, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "", + "lstrip": false, + "normalized": true, + "rstrip": false, + "single_word": false + }, + "pad_token": "", + "unk_token": { + "content": "", + "lstrip": false, + "normalized": true, + "rstrip": false, + "single_word": false + } +} diff --git a/tokenizer.model b/tokenizer.model new file mode 100644 index 0000000..6c00c74 --- /dev/null +++ b/tokenizer.model @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347 +size 499723 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..58c0f19 --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,43 @@ +{ + "add_bos_token": true, + "add_eos_token": false, + "added_tokens_decoder": { + "0": { + "content": "", + "lstrip": false, + "normalized": true, + "rstrip": false, + "single_word": false, + "special": true + }, + "1": { + "content": "", + "lstrip": false, + "normalized": true, + "rstrip": false, + "single_word": false, + "special": true + }, + "2": { + "content": "", + "lstrip": false, + "normalized": true, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "", + "clean_up_tokenization_spaces": false, + "eos_token": "", + "legacy": true, + "model_max_length": 2048, + "pad_token": "", + "padding_side": "right", + "sp_model_kwargs": {}, + "spaces_between_special_tokens": false, + "tokenizer_class": "LlamaTokenizer", + "unk_token": "", + "use_default_system_prompt": false, + "use_fast": true +} diff --git a/trainer_state.json b/trainer_state.json new file mode 100644 index 0000000..ca44c6e --- /dev/null +++ b/trainer_state.json @@ -0,0 +1,546 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.9980481457384514, + "eval_steps": 500, + "global_step": 4320, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.03, + "learning_rate": 7.692307692307694e-06, + "loss": 3.0678, + "step": 50 + }, + { + "epoch": 0.07, + "learning_rate": 1.5384615384615387e-05, + "loss": 2.3204, + "step": 100 + }, + { + "epoch": 0.1, + "learning_rate": 1.999887567062682e-05, + "loss": 1.9946, + "step": 150 + }, + { + "epoch": 0.14, + "learning_rate": 1.998622986851668e-05, + "loss": 1.8239, + "step": 200 + }, + { + "epoch": 0.17, + "learning_rate": 1.9959550682646195e-05, + "loss": 1.7682, + "step": 250 + }, + { + "epoch": 0.21, + "learning_rate": 1.9918875604567105e-05, + "loss": 1.6983, + "step": 300 + }, + { + "epoch": 0.24, + "learning_rate": 1.9864261793886955e-05, + "loss": 1.6675, + "step": 350 + }, + { + "epoch": 0.28, + "learning_rate": 1.9795785997944227e-05, + "loss": 1.6711, + "step": 400 + }, + { + "epoch": 0.31, + "learning_rate": 1.971354444395734e-05, + "loss": 1.6544, + "step": 450 + }, + { + "epoch": 0.35, + "learning_rate": 1.9617652703799102e-05, + "loss": 1.6135, + "step": 500 + }, + { + "epoch": 0.38, + "learning_rate": 1.9508245531586613e-05, + "loss": 1.6339, + "step": 550 + }, + { + "epoch": 0.42, + "learning_rate": 1.9385476674314883e-05, + "loss": 1.5796, + "step": 600 + }, + { + "epoch": 0.45, + "learning_rate": 1.9249518655800262e-05, + "loss": 1.5763, + "step": 650 + }, + { + "epoch": 0.49, + "learning_rate": 1.9100562534237318e-05, + "loss": 1.5615, + "step": 700 + }, + { + "epoch": 0.52, + "learning_rate": 1.8938817633709796e-05, + "loss": 1.5427, + "step": 750 + }, + { + "epoch": 0.56, + "learning_rate": 1.8764511250033053e-05, + "loss": 1.5316, + "step": 800 + }, + { + "epoch": 0.59, + "learning_rate": 1.8577888331341302e-05, + "loss": 1.5422, + "step": 850 + }, + { + "epoch": 0.62, + "learning_rate": 1.837921113386847e-05, + "loss": 1.5417, + "step": 900 + }, + { + "epoch": 0.66, + "learning_rate": 1.816875885340649e-05, + "loss": 1.5356, + "step": 950 + }, + { + "epoch": 0.69, + "learning_rate": 1.7946827232958858e-05, + "loss": 1.5343, + "step": 1000 + }, + { + "epoch": 0.73, + "learning_rate": 1.7713728147140857e-05, + "loss": 1.5125, + "step": 1050 + }, + { + "epoch": 0.76, + "learning_rate": 1.7469789163910416e-05, + "loss": 1.5159, + "step": 1100 + }, + { + "epoch": 0.8, + "learning_rate": 1.7215353084245564e-05, + "loss": 1.5107, + "step": 1150 + }, + { + "epoch": 0.83, + "learning_rate": 1.6950777460415324e-05, + "loss": 1.5229, + "step": 1200 + }, + { + "epoch": 0.87, + "learning_rate": 1.667643409352098e-05, + "loss": 1.4952, + "step": 1250 + }, + { + "epoch": 0.9, + "learning_rate": 1.6392708511013874e-05, + "loss": 1.5025, + "step": 1300 + }, + { + "epoch": 0.94, + "learning_rate": 1.6099999424923913e-05, + "loss": 1.488, + "step": 1350 + }, + { + "epoch": 0.97, + "learning_rate": 1.5798718171560124e-05, + "loss": 1.5125, + "step": 1400 + }, + { + "epoch": 1.01, + "learning_rate": 1.548928813347067e-05, + "loss": 1.506, + "step": 1450 + }, + { + "epoch": 1.04, + "learning_rate": 1.5172144144474582e-05, + "loss": 1.4895, + "step": 1500 + }, + { + "epoch": 1.08, + "learning_rate": 1.4847731878601348e-05, + "loss": 1.4411, + "step": 1550 + }, + { + "epoch": 1.11, + "learning_rate": 1.4516507223797018e-05, + "loss": 1.4307, + "step": 1600 + }, + { + "epoch": 1.15, + "learning_rate": 1.4178935641276982e-05, + "loss": 1.4289, + "step": 1650 + }, + { + "epoch": 1.18, + "learning_rate": 1.3835491511425678e-05, + "loss": 1.4333, + "step": 1700 + }, + { + "epoch": 1.21, + "learning_rate": 1.3486657467162423e-05, + "loss": 1.4518, + "step": 1750 + }, + { + "epoch": 1.25, + "learning_rate": 1.3132923715710191e-05, + "loss": 1.4258, + "step": 1800 + }, + { + "epoch": 1.28, + "learning_rate": 1.277478734972041e-05, + "loss": 1.4344, + "step": 1850 + }, + { + "epoch": 1.32, + "learning_rate": 1.2412751648721855e-05, + "loss": 1.434, + "step": 1900 + }, + { + "epoch": 1.35, + "learning_rate": 1.2047325371875286e-05, + "loss": 1.4473, + "step": 1950 + }, + { + "epoch": 1.39, + "learning_rate": 1.1679022043027706e-05, + "loss": 1.4234, + "step": 2000 + }, + { + "epoch": 1.42, + "learning_rate": 1.1308359229070914e-05, + "loss": 1.4298, + "step": 2050 + }, + { + "epoch": 1.46, + "learning_rate": 1.093585781261852e-05, + "loss": 1.4147, + "step": 2100 + }, + { + "epoch": 1.49, + "learning_rate": 1.0562041260023406e-05, + "loss": 1.4194, + "step": 2150 + }, + { + "epoch": 1.53, + "learning_rate": 1.018743488576437e-05, + "loss": 1.4264, + "step": 2200 + }, + { + "epoch": 1.56, + "learning_rate": 9.812565114235635e-06, + "loss": 1.4327, + "step": 2250 + }, + { + "epoch": 1.6, + "learning_rate": 9.437958739976597e-06, + "loss": 1.4231, + "step": 2300 + }, + { + "epoch": 1.63, + "learning_rate": 9.064142187381482e-06, + "loss": 1.397, + "step": 2350 + }, + { + "epoch": 1.67, + "learning_rate": 8.691640770929086e-06, + "loss": 1.4213, + "step": 2400 + }, + { + "epoch": 1.7, + "learning_rate": 8.3209779569723e-06, + "loss": 1.3827, + "step": 2450 + }, + { + "epoch": 1.73, + "learning_rate": 7.952674628124715e-06, + "loss": 1.4029, + "step": 2500 + }, + { + "epoch": 1.77, + "learning_rate": 7.587248351278149e-06, + "loss": 1.4026, + "step": 2550 + }, + { + "epoch": 1.8, + "learning_rate": 7.2252126502795935e-06, + "loss": 1.4032, + "step": 2600 + }, + { + "epoch": 1.84, + "learning_rate": 6.867076284289811e-06, + "loss": 1.4065, + "step": 2650 + }, + { + "epoch": 1.87, + "learning_rate": 6.513342532837581e-06, + "loss": 1.4033, + "step": 2700 + }, + { + "epoch": 1.91, + "learning_rate": 6.164508488574326e-06, + "loss": 1.4236, + "step": 2750 + }, + { + "epoch": 1.94, + "learning_rate": 5.821064358723021e-06, + "loss": 1.4105, + "step": 2800 + }, + { + "epoch": 1.98, + "learning_rate": 5.483492776202986e-06, + "loss": 1.4443, + "step": 2850 + }, + { + "epoch": 2.01, + "learning_rate": 5.152268121398656e-06, + "loss": 1.3914, + "step": 2900 + }, + { + "epoch": 2.05, + "learning_rate": 4.827855855525423e-06, + "loss": 1.3694, + "step": 2950 + }, + { + "epoch": 2.08, + "learning_rate": 4.510711866529335e-06, + "loss": 1.3685, + "step": 3000 + }, + { + "epoch": 2.12, + "learning_rate": 4.201281828439878e-06, + "loss": 1.38, + "step": 3050 + }, + { + "epoch": 2.15, + "learning_rate": 3.9000005750760915e-06, + "loss": 1.3944, + "step": 3100 + }, + { + "epoch": 2.19, + "learning_rate": 3.6072914889861287e-06, + "loss": 1.3632, + "step": 3150 + }, + { + "epoch": 2.22, + "learning_rate": 3.323565906479024e-06, + "loss": 1.3923, + "step": 3200 + }, + { + "epoch": 2.26, + "learning_rate": 3.0492225395846777e-06, + "loss": 1.3598, + "step": 3250 + }, + { + "epoch": 2.29, + "learning_rate": 2.7846469157544353e-06, + "loss": 1.3582, + "step": 3300 + }, + { + "epoch": 2.32, + "learning_rate": 2.530210836089586e-06, + "loss": 1.4019, + "step": 3350 + }, + { + "epoch": 2.36, + "learning_rate": 2.2862718528591434e-06, + "loss": 1.3822, + "step": 3400 + }, + { + "epoch": 2.39, + "learning_rate": 2.0531727670411418e-06, + "loss": 1.4255, + "step": 3450 + }, + { + "epoch": 2.43, + "learning_rate": 1.831241146593512e-06, + "loss": 1.3501, + "step": 3500 + }, + { + "epoch": 2.46, + "learning_rate": 1.620788866131533e-06, + "loss": 1.3792, + "step": 3550 + }, + { + "epoch": 2.5, + "learning_rate": 1.4221116686587e-06, + "loss": 1.3632, + "step": 3600 + }, + { + "epoch": 2.53, + "learning_rate": 1.2354887499669487e-06, + "loss": 1.389, + "step": 3650 + }, + { + "epoch": 2.57, + "learning_rate": 1.0611823662902088e-06, + "loss": 1.3764, + "step": 3700 + }, + { + "epoch": 2.6, + "learning_rate": 8.994374657626847e-07, + "loss": 1.3873, + "step": 3750 + }, + { + "epoch": 2.64, + "learning_rate": 7.504813441997382e-07, + "loss": 1.3481, + "step": 3800 + }, + { + "epoch": 2.67, + "learning_rate": 6.145233256851202e-07, + "loss": 1.3685, + "step": 3850 + }, + { + "epoch": 2.71, + "learning_rate": 4.91754468413388e-07, + "loss": 1.3712, + "step": 3900 + }, + { + "epoch": 2.74, + "learning_rate": 3.823472962008978e-07, + "loss": 1.3702, + "step": 3950 + }, + { + "epoch": 2.78, + "learning_rate": 2.864555560426596e-07, + "loss": 1.3731, + "step": 4000 + }, + { + "epoch": 2.81, + "learning_rate": 2.0421400205577591e-07, + "loss": 1.392, + "step": 4050 + }, + { + "epoch": 2.85, + "learning_rate": 1.3573820611304745e-07, + "loss": 1.3628, + "step": 4100 + }, + { + "epoch": 2.88, + "learning_rate": 8.112439543289663e-08, + "loss": 1.367, + "step": 4150 + }, + { + "epoch": 2.91, + "learning_rate": 4.044931735380564e-08, + "loss": 1.371, + "step": 4200 + }, + { + "epoch": 2.95, + "learning_rate": 1.3770131483322734e-08, + "loss": 1.3728, + "step": 4250 + }, + { + "epoch": 2.98, + "learning_rate": 1.1243293731799665e-09, + "loss": 1.3754, + "step": 4300 + }, + { + "epoch": 3.0, + "step": 4320, + "total_flos": 4.683004912987013e+17, + "train_loss": 1.4905582423563357, + "train_runtime": 19585.3727, + "train_samples_per_second": 14.126, + "train_steps_per_second": 0.221 + } + ], + "logging_steps": 50, + "max_steps": 4320, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 1200, + "total_flos": 4.683004912987013e+17, + "train_batch_size": 2, + "trial_name": null, + "trial_params": null +} diff --git a/training_args.bin b/training_args.bin new file mode 100644 index 0000000..f584e7a --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f89e6d423314f571519b58181726e976526e530a845a65d710f27578736c0222 +size 4283