diff --git a/multi-prompt/twochar/README.md b/multi-prompt/twochar/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/multi-prompt/twochar/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/multi-prompt/twochar/adapter_config.json b/multi-prompt/twochar/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b33f8685004fc9f9eef2843be41eb6081508fc38 --- /dev/null +++ b/multi-prompt/twochar/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.01, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "v_proj", + "q_proj", + "down_proj", + "k_proj", + "gate_proj", + "o_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/multi-prompt/twochar/adapter_model.bin b/multi-prompt/twochar/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..c79fe6f2dcda218f8075f5ac37aaf3f0caeff3d8 --- /dev/null +++ b/multi-prompt/twochar/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:433a53226b2a4581547b8a2e381d865d58aa0575acf593735501071cf46d7fef +size 125374989 diff --git a/multi-prompt/twochar/added_tokens.json b/multi-prompt/twochar/added_tokens.json new file mode 100644 index 0000000000000000000000000000000000000000..e41416ddd79948246ea2dced6800ea3cd531c424 --- /dev/null +++ b/multi-prompt/twochar/added_tokens.json @@ -0,0 +1,3 @@ +{ + "[PAD]": 32000 +} diff --git a/multi-prompt/twochar/checkpoint-100/README.md b/multi-prompt/twochar/checkpoint-100/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/multi-prompt/twochar/checkpoint-100/adapter_config.json b/multi-prompt/twochar/checkpoint-100/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b33f8685004fc9f9eef2843be41eb6081508fc38 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.01, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "v_proj", + "q_proj", + "down_proj", + "k_proj", + "gate_proj", + "o_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/multi-prompt/twochar/checkpoint-100/adapter_model.bin b/multi-prompt/twochar/checkpoint-100/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..e81a2b2e0a25412c7d260c0d5d2c6a06f2e69f19 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0708b4cbcf4d2eab6bb13a3b593efb8f9ffd0d3554bd6f0d85be7c2443d26b80 +size 125374989 diff --git a/multi-prompt/twochar/checkpoint-100/adapter_model/README.md b/multi-prompt/twochar/checkpoint-100/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/adapter_model/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/multi-prompt/twochar/checkpoint-100/adapter_model/adapter_config.json b/multi-prompt/twochar/checkpoint-100/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b33f8685004fc9f9eef2843be41eb6081508fc38 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/adapter_model/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.01, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "v_proj", + "q_proj", + "down_proj", + "k_proj", + "gate_proj", + "o_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/multi-prompt/twochar/checkpoint-100/adapter_model/adapter_model.bin b/multi-prompt/twochar/checkpoint-100/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..e81a2b2e0a25412c7d260c0d5d2c6a06f2e69f19 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0708b4cbcf4d2eab6bb13a3b593efb8f9ffd0d3554bd6f0d85be7c2443d26b80 +size 125374989 diff --git a/multi-prompt/twochar/checkpoint-100/optimizer.pt b/multi-prompt/twochar/checkpoint-100/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..aaf808f10c0b4118259cb0a33780917be0e14042 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0221233935a3705a400234359b0d51ecb64e96906b36ad8742d87e4e81167bd8 +size 250681597 diff --git a/multi-prompt/twochar/checkpoint-100/rng_state_0.pth b/multi-prompt/twochar/checkpoint-100/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..c027d7e1b409ca7399334f4fb94eb2145805c56d --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef3b38ac0e685cf7486af95e3e44d65bd9d5f0166e4b7cda4087f3df177fe2ac +size 21687 diff --git a/multi-prompt/twochar/checkpoint-100/rng_state_1.pth b/multi-prompt/twochar/checkpoint-100/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..4365f610fe476feb9e4a0815636c0f03dc780878 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:96f602cf80bbe8b7b20f61815b81a180e05c4349767b3055d38cee11e56d2f52 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-100/rng_state_2.pth b/multi-prompt/twochar/checkpoint-100/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..e78b1d9b263773f1fc17c537a59fccaad3fa4237 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c4c642ebadc5d5d3863c2af80d71300c9b960385b08a7e97d47221351e923536 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-100/rng_state_3.pth b/multi-prompt/twochar/checkpoint-100/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..e18b7bd48ce79cdea57bade728fd0ee5570b5300 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bdbab4ca5a74f90e77ec13e5c8a300b12d2ec0ffa5f7150ef45db3b39c0ba3fd +size 21687 diff --git a/multi-prompt/twochar/checkpoint-100/rng_state_4.pth b/multi-prompt/twochar/checkpoint-100/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..686c2ba0482c0a3291c1c693317c281ba57c82a0 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8b9e295735ec1efb6e545052a0a5b42c962567ac43c037ef05cd67f65aac86e5 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-100/rng_state_5.pth b/multi-prompt/twochar/checkpoint-100/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..d30c7d2e3633c5923fbec2dd60c0959aa47b1699 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bf0b7826ae8f9d6847c100c005a195903f6702ffefbcf20043e504db1cbb9cf0 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-100/rng_state_6.pth b/multi-prompt/twochar/checkpoint-100/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..580b669b95dd45546196ccdc669d2b2365f7ba8e --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c9b6a36344bf4dd837aca81f6f221eaaf42fdaf4ad38def06c6583bd294478b +size 21687 diff --git a/multi-prompt/twochar/checkpoint-100/rng_state_7.pth b/multi-prompt/twochar/checkpoint-100/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..b469c82317c3fbe5c8f9d58b23b976eeac1777d4 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fe85df90b682b7f8f20f078de75c81ecf060770f57b3f4465e9d00985eaa8a39 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-100/scheduler.pt b/multi-prompt/twochar/checkpoint-100/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..36e20fc64cdb8ac5215ac226a0de4079c0dc732a --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b8c8c5e17820ade3bd0f83be62a888f354614fa9e9626e94814a99519af6a262 +size 627 diff --git a/multi-prompt/twochar/checkpoint-100/trainer_state.json b/multi-prompt/twochar/checkpoint-100/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..161723e07a2e1943ff3ec6ff69406115e373d4c9 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/trainer_state.json @@ -0,0 +1,659 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.9900990099009901, + "eval_steps": 20, + "global_step": 100, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.01, + "learning_rate": 6.5e-06, + "loss": 2.0407, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 1.3e-05, + "loss": 2.3107, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 1.9499999999999996e-05, + "loss": 2.1613, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 2.6e-05, + "loss": 2.0444, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 3.25e-05, + "loss": 2.2512, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 3.899999999999999e-05, + "loss": 2.0516, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 4.5499999999999995e-05, + "loss": 2.2006, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 5.2e-05, + "loss": 2.1229, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 5.85e-05, + "loss": 2.2498, + "step": 9 + }, + { + "epoch": 0.1, + "learning_rate": 6.5e-05, + "loss": 2.0293, + "step": 10 + }, + { + "epoch": 0.11, + "learning_rate": 6.499564948206075e-05, + "loss": 2.1316, + "step": 11 + }, + { + "epoch": 0.12, + "learning_rate": 6.498259909298188e-05, + "loss": 2.1274, + "step": 12 + }, + { + "epoch": 0.13, + "learning_rate": 6.49608523266681e-05, + "loss": 2.1129, + "step": 13 + }, + { + "epoch": 0.14, + "learning_rate": 6.493041500525461e-05, + "loss": 2.1517, + "step": 14 + }, + { + "epoch": 0.15, + "learning_rate": 6.489129527754836e-05, + "loss": 2.1078, + "step": 15 + }, + { + "epoch": 0.16, + "learning_rate": 6.484350361684639e-05, + "loss": 2.0362, + "step": 16 + }, + { + "epoch": 0.17, + "learning_rate": 6.478705281813194e-05, + "loss": 2.0553, + "step": 17 + }, + { + "epoch": 0.18, + "learning_rate": 6.472195799464884e-05, + "loss": 2.1223, + "step": 18 + }, + { + "epoch": 0.19, + "learning_rate": 6.464823657385538e-05, + "loss": 2.0326, + "step": 19 + }, + { + "epoch": 0.2, + "learning_rate": 6.456590829275857e-05, + "loss": 2.1802, + "step": 20 + }, + { + "epoch": 0.2, + "eval_loss": 2.012301445007324, + "eval_runtime": 1.4855, + "eval_samples_per_second": 5.385, + "eval_steps_per_second": 0.673, + "step": 20 + }, + { + "epoch": 0.21, + "learning_rate": 6.447499519263001e-05, + "loss": 2.0665, + "step": 21 + }, + { + "epoch": 0.22, + "learning_rate": 6.437552161310498e-05, + "loss": 2.0354, + "step": 22 + }, + { + "epoch": 0.23, + "learning_rate": 6.426751418566609e-05, + "loss": 2.1289, + "step": 23 + }, + { + "epoch": 0.24, + "learning_rate": 6.415100182651334e-05, + "loss": 2.0742, + "step": 24 + }, + { + "epoch": 0.25, + "learning_rate": 6.402601572882268e-05, + "loss": 2.0412, + "step": 25 + }, + { + "epoch": 0.26, + "learning_rate": 6.38925893543947e-05, + "loss": 2.1974, + "step": 26 + }, + { + "epoch": 0.27, + "learning_rate": 6.375075842469626e-05, + "loss": 2.196, + "step": 27 + }, + { + "epoch": 0.28, + "learning_rate": 6.360056091129678e-05, + "loss": 2.0775, + "step": 28 + }, + { + "epoch": 0.29, + "learning_rate": 6.344203702570254e-05, + "loss": 2.0366, + "step": 29 + }, + { + "epoch": 0.3, + "learning_rate": 6.327522920859093e-05, + "loss": 2.1438, + "step": 30 + }, + { + "epoch": 0.31, + "learning_rate": 6.310018211844817e-05, + "loss": 2.0639, + "step": 31 + }, + { + "epoch": 0.32, + "learning_rate": 6.291694261961308e-05, + "loss": 2.1794, + "step": 32 + }, + { + "epoch": 0.33, + "learning_rate": 6.272555976973039e-05, + "loss": 2.0592, + "step": 33 + }, + { + "epoch": 0.34, + "learning_rate": 6.252608480661681e-05, + "loss": 2.1628, + "step": 34 + }, + { + "epoch": 0.35, + "learning_rate": 6.231857113454346e-05, + "loss": 2.1851, + "step": 35 + }, + { + "epoch": 0.36, + "learning_rate": 6.21030743099382e-05, + "loss": 2.0531, + "step": 36 + }, + { + "epoch": 0.37, + "learning_rate": 6.18796520265119e-05, + "loss": 2.0284, + "step": 37 + }, + { + "epoch": 0.38, + "learning_rate": 6.164836409981237e-05, + "loss": 2.2359, + "step": 38 + }, + { + "epoch": 0.39, + "learning_rate": 6.140927245121032e-05, + "loss": 1.8967, + "step": 39 + }, + { + "epoch": 0.4, + "learning_rate": 6.116244109132153e-05, + "loss": 2.0275, + "step": 40 + }, + { + "epoch": 0.4, + "eval_loss": 1.9861761331558228, + "eval_runtime": 1.4896, + "eval_samples_per_second": 5.37, + "eval_steps_per_second": 0.671, + "step": 40 + }, + { + "epoch": 0.41, + "learning_rate": 6.0907936102869656e-05, + "loss": 2.0316, + "step": 41 + }, + { + "epoch": 0.42, + "learning_rate": 6.0645825622994254e-05, + "loss": 2.0768, + "step": 42 + }, + { + "epoch": 0.43, + "learning_rate": 6.037617982500884e-05, + "loss": 2.0058, + "step": 43 + }, + { + "epoch": 0.44, + "learning_rate": 6.009907089961381e-05, + "loss": 2.2319, + "step": 44 + }, + { + "epoch": 0.45, + "learning_rate": 5.981457303556923e-05, + "loss": 2.1693, + "step": 45 + }, + { + "epoch": 0.46, + "learning_rate": 5.9522762399832716e-05, + "loss": 2.0858, + "step": 46 + }, + { + "epoch": 0.47, + "learning_rate": 5.922371711716768e-05, + "loss": 2.0576, + "step": 47 + }, + { + "epoch": 0.48, + "learning_rate": 5.891751724922749e-05, + "loss": 1.9946, + "step": 48 + }, + { + "epoch": 0.49, + "learning_rate": 5.860424477312095e-05, + "loss": 2.0166, + "step": 49 + }, + { + "epoch": 0.5, + "learning_rate": 5.828398355946514e-05, + "loss": 2.1482, + "step": 50 + }, + { + "epoch": 0.5, + "learning_rate": 5.795681934993113e-05, + "loss": 2.0698, + "step": 51 + }, + { + "epoch": 0.51, + "learning_rate": 5.7622839734288945e-05, + "loss": 1.9317, + "step": 52 + }, + { + "epoch": 0.52, + "learning_rate": 5.728213412695761e-05, + "loss": 2.1399, + "step": 53 + }, + { + "epoch": 0.53, + "learning_rate": 5.693479374306676e-05, + "loss": 2.1637, + "step": 54 + }, + { + "epoch": 0.54, + "learning_rate": 5.658091157403617e-05, + "loss": 2.0557, + "step": 55 + }, + { + "epoch": 0.55, + "learning_rate": 5.622058236267965e-05, + "loss": 1.926, + "step": 56 + }, + { + "epoch": 0.56, + "learning_rate": 5.585390257784018e-05, + "loss": 2.0984, + "step": 57 + }, + { + "epoch": 0.57, + "learning_rate": 5.548097038856279e-05, + "loss": 2.0428, + "step": 58 + }, + { + "epoch": 0.58, + "learning_rate": 5.5101885637812374e-05, + "loss": 1.9624, + "step": 59 + }, + { + "epoch": 0.59, + "learning_rate": 5.4716749815743304e-05, + "loss": 1.913, + "step": 60 + }, + { + "epoch": 0.59, + "eval_loss": 1.97765052318573, + "eval_runtime": 1.4886, + "eval_samples_per_second": 5.374, + "eval_steps_per_second": 0.672, + "step": 60 + }, + { + "epoch": 0.6, + "learning_rate": 5.432566603252809e-05, + "loss": 2.0125, + "step": 61 + }, + { + "epoch": 0.61, + "learning_rate": 5.3928738990752234e-05, + "loss": 2.0721, + "step": 62 + }, + { + "epoch": 0.62, + "learning_rate": 5.3526074957382866e-05, + "loss": 2.1046, + "step": 63 + }, + { + "epoch": 0.63, + "learning_rate": 5.311778173531847e-05, + "loss": 2.0074, + "step": 64 + }, + { + "epoch": 0.64, + "learning_rate": 5.2703968634527514e-05, + "loss": 2.1667, + "step": 65 + }, + { + "epoch": 0.65, + "learning_rate": 5.2284746442783414e-05, + "loss": 2.0677, + "step": 66 + }, + { + "epoch": 0.66, + "learning_rate": 5.186022739600408e-05, + "loss": 2.1798, + "step": 67 + }, + { + "epoch": 0.67, + "learning_rate": 5.143052514820357e-05, + "loss": 2.105, + "step": 68 + }, + { + "epoch": 0.68, + "learning_rate": 5.099575474106419e-05, + "loss": 2.118, + "step": 69 + }, + { + "epoch": 0.69, + "learning_rate": 5.055603257313707e-05, + "loss": 2.0791, + "step": 70 + }, + { + "epoch": 0.7, + "learning_rate": 5.011147636867943e-05, + "loss": 2.032, + "step": 71 + }, + { + "epoch": 0.71, + "learning_rate": 4.9662205146136955e-05, + "loss": 1.9331, + "step": 72 + }, + { + "epoch": 0.72, + "learning_rate": 4.92083391862797e-05, + "loss": 2.0077, + "step": 73 + }, + { + "epoch": 0.73, + "learning_rate": 4.874999999999999e-05, + "loss": 2.0253, + "step": 74 + }, + { + "epoch": 0.74, + "learning_rate": 4.828731029578105e-05, + "loss": 1.9922, + "step": 75 + }, + { + "epoch": 0.75, + "learning_rate": 4.7820393946844926e-05, + "loss": 2.0295, + "step": 76 + }, + { + "epoch": 0.76, + "learning_rate": 4.734937595798867e-05, + "loss": 1.9514, + "step": 77 + }, + { + "epoch": 0.77, + "learning_rate": 4.687438243211754e-05, + "loss": 2.1018, + "step": 78 + }, + { + "epoch": 0.78, + "learning_rate": 4.639554053648416e-05, + "loss": 2.1032, + "step": 79 + }, + { + "epoch": 0.79, + "learning_rate": 4.5912978468642824e-05, + "loss": 2.1258, + "step": 80 + }, + { + "epoch": 0.79, + "eval_loss": 1.972991704940796, + "eval_runtime": 1.494, + "eval_samples_per_second": 5.355, + "eval_steps_per_second": 0.669, + "step": 80 + }, + { + "epoch": 0.8, + "learning_rate": 4.542682542212785e-05, + "loss": 2.0353, + "step": 81 + }, + { + "epoch": 0.81, + "learning_rate": 4.4937211551865415e-05, + "loss": 1.9987, + "step": 82 + }, + { + "epoch": 0.82, + "learning_rate": 4.444426793932787e-05, + "loss": 1.8856, + "step": 83 + }, + { + "epoch": 0.83, + "learning_rate": 4.3948126557440085e-05, + "loss": 2.0904, + "step": 84 + }, + { + "epoch": 0.84, + "learning_rate": 4.344892023524714e-05, + "loss": 2.0685, + "step": 85 + }, + { + "epoch": 0.85, + "learning_rate": 4.2946782622352746e-05, + "loss": 2.1, + "step": 86 + }, + { + "epoch": 0.86, + "learning_rate": 4.244184815313799e-05, + "loss": 1.9322, + "step": 87 + }, + { + "epoch": 0.87, + "learning_rate": 4.193425201077002e-05, + "loss": 2.194, + "step": 88 + }, + { + "epoch": 0.88, + "learning_rate": 4.14241300910103e-05, + "loss": 2.0455, + "step": 89 + }, + { + "epoch": 0.89, + "learning_rate": 4.091161896583192e-05, + "loss": 2.1239, + "step": 90 + }, + { + "epoch": 0.9, + "learning_rate": 4.0396855846856076e-05, + "loss": 2.142, + "step": 91 + }, + { + "epoch": 0.91, + "learning_rate": 3.9879978548617125e-05, + "loss": 2.2304, + "step": 92 + }, + { + "epoch": 0.92, + "learning_rate": 3.936112545166636e-05, + "loss": 1.925, + "step": 93 + }, + { + "epoch": 0.93, + "learning_rate": 3.884043546552417e-05, + "loss": 2.1547, + "step": 94 + }, + { + "epoch": 0.94, + "learning_rate": 3.831804799149057e-05, + "loss": 2.0439, + "step": 95 + }, + { + "epoch": 0.95, + "learning_rate": 3.779410288532413e-05, + "loss": 1.9763, + "step": 96 + }, + { + "epoch": 0.96, + "learning_rate": 3.726874041979925e-05, + "loss": 2.114, + "step": 97 + }, + { + "epoch": 0.97, + "learning_rate": 3.674210124715168e-05, + "loss": 2.0293, + "step": 98 + }, + { + "epoch": 0.98, + "learning_rate": 3.621432636142251e-05, + "loss": 2.0833, + "step": 99 + }, + { + "epoch": 0.99, + "learning_rate": 3.5685557060710726e-05, + "loss": 2.0452, + "step": 100 + }, + { + "epoch": 0.99, + "eval_loss": 1.9682413339614868, + "eval_runtime": 1.4934, + "eval_samples_per_second": 5.357, + "eval_steps_per_second": 0.67, + "step": 100 + } + ], + "logging_steps": 1, + "max_steps": 202, + "num_train_epochs": 2, + "save_steps": 50, + "total_flos": 8.359599153348608e+16, + "trial_name": null, + "trial_params": null +} diff --git a/multi-prompt/twochar/checkpoint-100/training_args.bin b/multi-prompt/twochar/checkpoint-100/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..47d6200b8d4047ff364e6b15cf014dde61f2166b --- /dev/null +++ b/multi-prompt/twochar/checkpoint-100/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac3fb199292750362f1e387a5d8bde7088e81478bbe173194e512403378f1c70 +size 4475 diff --git a/multi-prompt/twochar/checkpoint-150/README.md b/multi-prompt/twochar/checkpoint-150/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/multi-prompt/twochar/checkpoint-150/adapter_config.json b/multi-prompt/twochar/checkpoint-150/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b33f8685004fc9f9eef2843be41eb6081508fc38 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.01, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "v_proj", + "q_proj", + "down_proj", + "k_proj", + "gate_proj", + "o_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/multi-prompt/twochar/checkpoint-150/adapter_model.bin b/multi-prompt/twochar/checkpoint-150/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..4f8ca6891c7773372d69fb2ad8df2ffaa55debb5 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:94befd076d47fa24e3ba916e9a3778dd8ece4972bb7c203083cc5f53c5e46bfa +size 125374989 diff --git a/multi-prompt/twochar/checkpoint-150/adapter_model/README.md b/multi-prompt/twochar/checkpoint-150/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/adapter_model/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/multi-prompt/twochar/checkpoint-150/adapter_model/adapter_config.json b/multi-prompt/twochar/checkpoint-150/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b33f8685004fc9f9eef2843be41eb6081508fc38 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/adapter_model/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.01, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "v_proj", + "q_proj", + "down_proj", + "k_proj", + "gate_proj", + "o_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/multi-prompt/twochar/checkpoint-150/adapter_model/adapter_model.bin b/multi-prompt/twochar/checkpoint-150/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..4f8ca6891c7773372d69fb2ad8df2ffaa55debb5 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:94befd076d47fa24e3ba916e9a3778dd8ece4972bb7c203083cc5f53c5e46bfa +size 125374989 diff --git a/multi-prompt/twochar/checkpoint-150/optimizer.pt b/multi-prompt/twochar/checkpoint-150/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..f561be21d8cb9043f5dda0d68fc3f76b25f632f3 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:60fd84426c8963b86174be0513aea26512fafa4a79de37a27d66e42699c92c1d +size 250681597 diff --git a/multi-prompt/twochar/checkpoint-150/rng_state_0.pth b/multi-prompt/twochar/checkpoint-150/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..2e852bd74a8f93ab1c4de0f44917d87779fa6280 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d127d4391ee6771b6d20db8ce3f0dd0584c753057cc9bdfdd0be6f8cef84d955 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-150/rng_state_1.pth b/multi-prompt/twochar/checkpoint-150/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..967c3b57c88f0fd6224fc782325081203d780c53 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d7ccf1beada17255706986caf9094f244684163a8547f833874af8de8dd4daa0 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-150/rng_state_2.pth b/multi-prompt/twochar/checkpoint-150/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..01a3800c419d7a10b3033798c115a4599ca65708 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:277501f7b2a65d7470e01413187a94540bcb41c51790d63ac495be26b8fdfb89 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-150/rng_state_3.pth b/multi-prompt/twochar/checkpoint-150/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..3f49ae80b4f69a09a48a97d4fab2e4dc9306d4b0 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1d55ef5ebc7e10e4cd63d908d65f05dfdc0de99fd4911bc3aa2e57e01940ef54 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-150/rng_state_4.pth b/multi-prompt/twochar/checkpoint-150/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..53096add211f9c568abe4840b3309d8fa88cf191 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ddc9393cb55ab7d202c2cc2ae933435652c478277daf312e6370eb3404db7e9f +size 21687 diff --git a/multi-prompt/twochar/checkpoint-150/rng_state_5.pth b/multi-prompt/twochar/checkpoint-150/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..61804cccdfd8c946e52825a7b57b5c48feb69634 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fcbf8760f23fffddb8debdb97db5712925a0c049476c45b8b57488cb099ff82e +size 21687 diff --git a/multi-prompt/twochar/checkpoint-150/rng_state_6.pth b/multi-prompt/twochar/checkpoint-150/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..d0be26302b066358e34db48b75ac1314b1a7dc78 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:db101559fa9579ad6b9ca8b4a109575c11fc03a27ef55f3e0bb92c5642fc4001 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-150/rng_state_7.pth b/multi-prompt/twochar/checkpoint-150/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..26ace6a6033c3e5cabf8c074559423187fe8f8d5 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:644bacb2bf138ffa7d1d20a3358a4fab98d6ab8c6ede06df95e5de696a31d2ce +size 21687 diff --git a/multi-prompt/twochar/checkpoint-150/scheduler.pt b/multi-prompt/twochar/checkpoint-150/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..239565dc0a5d8045d4945debf9c17e89b35c6b1f --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c4e06f3498f4a96af42d371e2251f32c6afc73dc33f11f9b431ef7f76337ee64 +size 627 diff --git a/multi-prompt/twochar/checkpoint-150/trainer_state.json b/multi-prompt/twochar/checkpoint-150/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..027041939045976d83cbeaf08797e3bda1e610de --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/trainer_state.json @@ -0,0 +1,975 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.4851485148514851, + "eval_steps": 20, + "global_step": 150, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.01, + "learning_rate": 6.5e-06, + "loss": 2.0407, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 1.3e-05, + "loss": 2.3107, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 1.9499999999999996e-05, + "loss": 2.1613, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 2.6e-05, + "loss": 2.0444, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 3.25e-05, + "loss": 2.2512, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 3.899999999999999e-05, + "loss": 2.0516, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 4.5499999999999995e-05, + "loss": 2.2006, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 5.2e-05, + "loss": 2.1229, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 5.85e-05, + "loss": 2.2498, + "step": 9 + }, + { + "epoch": 0.1, + "learning_rate": 6.5e-05, + "loss": 2.0293, + "step": 10 + }, + { + "epoch": 0.11, + "learning_rate": 6.499564948206075e-05, + "loss": 2.1316, + "step": 11 + }, + { + "epoch": 0.12, + "learning_rate": 6.498259909298188e-05, + "loss": 2.1274, + "step": 12 + }, + { + "epoch": 0.13, + "learning_rate": 6.49608523266681e-05, + "loss": 2.1129, + "step": 13 + }, + { + "epoch": 0.14, + "learning_rate": 6.493041500525461e-05, + "loss": 2.1517, + "step": 14 + }, + { + "epoch": 0.15, + "learning_rate": 6.489129527754836e-05, + "loss": 2.1078, + "step": 15 + }, + { + "epoch": 0.16, + "learning_rate": 6.484350361684639e-05, + "loss": 2.0362, + "step": 16 + }, + { + "epoch": 0.17, + "learning_rate": 6.478705281813194e-05, + "loss": 2.0553, + "step": 17 + }, + { + "epoch": 0.18, + "learning_rate": 6.472195799464884e-05, + "loss": 2.1223, + "step": 18 + }, + { + "epoch": 0.19, + "learning_rate": 6.464823657385538e-05, + "loss": 2.0326, + "step": 19 + }, + { + "epoch": 0.2, + "learning_rate": 6.456590829275857e-05, + "loss": 2.1802, + "step": 20 + }, + { + "epoch": 0.2, + "eval_loss": 2.012301445007324, + "eval_runtime": 1.4855, + "eval_samples_per_second": 5.385, + "eval_steps_per_second": 0.673, + "step": 20 + }, + { + "epoch": 0.21, + "learning_rate": 6.447499519263001e-05, + "loss": 2.0665, + "step": 21 + }, + { + "epoch": 0.22, + "learning_rate": 6.437552161310498e-05, + "loss": 2.0354, + "step": 22 + }, + { + "epoch": 0.23, + "learning_rate": 6.426751418566609e-05, + "loss": 2.1289, + "step": 23 + }, + { + "epoch": 0.24, + "learning_rate": 6.415100182651334e-05, + "loss": 2.0742, + "step": 24 + }, + { + "epoch": 0.25, + "learning_rate": 6.402601572882268e-05, + "loss": 2.0412, + "step": 25 + }, + { + "epoch": 0.26, + "learning_rate": 6.38925893543947e-05, + "loss": 2.1974, + "step": 26 + }, + { + "epoch": 0.27, + "learning_rate": 6.375075842469626e-05, + "loss": 2.196, + "step": 27 + }, + { + "epoch": 0.28, + "learning_rate": 6.360056091129678e-05, + "loss": 2.0775, + "step": 28 + }, + { + "epoch": 0.29, + "learning_rate": 6.344203702570254e-05, + "loss": 2.0366, + "step": 29 + }, + { + "epoch": 0.3, + "learning_rate": 6.327522920859093e-05, + "loss": 2.1438, + "step": 30 + }, + { + "epoch": 0.31, + "learning_rate": 6.310018211844817e-05, + "loss": 2.0639, + "step": 31 + }, + { + "epoch": 0.32, + "learning_rate": 6.291694261961308e-05, + "loss": 2.1794, + "step": 32 + }, + { + "epoch": 0.33, + "learning_rate": 6.272555976973039e-05, + "loss": 2.0592, + "step": 33 + }, + { + "epoch": 0.34, + "learning_rate": 6.252608480661681e-05, + "loss": 2.1628, + "step": 34 + }, + { + "epoch": 0.35, + "learning_rate": 6.231857113454346e-05, + "loss": 2.1851, + "step": 35 + }, + { + "epoch": 0.36, + "learning_rate": 6.21030743099382e-05, + "loss": 2.0531, + "step": 36 + }, + { + "epoch": 0.37, + "learning_rate": 6.18796520265119e-05, + "loss": 2.0284, + "step": 37 + }, + { + "epoch": 0.38, + "learning_rate": 6.164836409981237e-05, + "loss": 2.2359, + "step": 38 + }, + { + "epoch": 0.39, + "learning_rate": 6.140927245121032e-05, + "loss": 1.8967, + "step": 39 + }, + { + "epoch": 0.4, + "learning_rate": 6.116244109132153e-05, + "loss": 2.0275, + "step": 40 + }, + { + "epoch": 0.4, + "eval_loss": 1.9861761331558228, + "eval_runtime": 1.4896, + "eval_samples_per_second": 5.37, + "eval_steps_per_second": 0.671, + "step": 40 + }, + { + "epoch": 0.41, + "learning_rate": 6.0907936102869656e-05, + "loss": 2.0316, + "step": 41 + }, + { + "epoch": 0.42, + "learning_rate": 6.0645825622994254e-05, + "loss": 2.0768, + "step": 42 + }, + { + "epoch": 0.43, + "learning_rate": 6.037617982500884e-05, + "loss": 2.0058, + "step": 43 + }, + { + "epoch": 0.44, + "learning_rate": 6.009907089961381e-05, + "loss": 2.2319, + "step": 44 + }, + { + "epoch": 0.45, + "learning_rate": 5.981457303556923e-05, + "loss": 2.1693, + "step": 45 + }, + { + "epoch": 0.46, + "learning_rate": 5.9522762399832716e-05, + "loss": 2.0858, + "step": 46 + }, + { + "epoch": 0.47, + "learning_rate": 5.922371711716768e-05, + "loss": 2.0576, + "step": 47 + }, + { + "epoch": 0.48, + "learning_rate": 5.891751724922749e-05, + "loss": 1.9946, + "step": 48 + }, + { + "epoch": 0.49, + "learning_rate": 5.860424477312095e-05, + "loss": 2.0166, + "step": 49 + }, + { + "epoch": 0.5, + "learning_rate": 5.828398355946514e-05, + "loss": 2.1482, + "step": 50 + }, + { + "epoch": 0.5, + "learning_rate": 5.795681934993113e-05, + "loss": 2.0698, + "step": 51 + }, + { + "epoch": 0.51, + "learning_rate": 5.7622839734288945e-05, + "loss": 1.9317, + "step": 52 + }, + { + "epoch": 0.52, + "learning_rate": 5.728213412695761e-05, + "loss": 2.1399, + "step": 53 + }, + { + "epoch": 0.53, + "learning_rate": 5.693479374306676e-05, + "loss": 2.1637, + "step": 54 + }, + { + "epoch": 0.54, + "learning_rate": 5.658091157403617e-05, + "loss": 2.0557, + "step": 55 + }, + { + "epoch": 0.55, + "learning_rate": 5.622058236267965e-05, + "loss": 1.926, + "step": 56 + }, + { + "epoch": 0.56, + "learning_rate": 5.585390257784018e-05, + "loss": 2.0984, + "step": 57 + }, + { + "epoch": 0.57, + "learning_rate": 5.548097038856279e-05, + "loss": 2.0428, + "step": 58 + }, + { + "epoch": 0.58, + "learning_rate": 5.5101885637812374e-05, + "loss": 1.9624, + "step": 59 + }, + { + "epoch": 0.59, + "learning_rate": 5.4716749815743304e-05, + "loss": 1.913, + "step": 60 + }, + { + "epoch": 0.59, + "eval_loss": 1.97765052318573, + "eval_runtime": 1.4886, + "eval_samples_per_second": 5.374, + "eval_steps_per_second": 0.672, + "step": 60 + }, + { + "epoch": 0.6, + "learning_rate": 5.432566603252809e-05, + "loss": 2.0125, + "step": 61 + }, + { + "epoch": 0.61, + "learning_rate": 5.3928738990752234e-05, + "loss": 2.0721, + "step": 62 + }, + { + "epoch": 0.62, + "learning_rate": 5.3526074957382866e-05, + "loss": 2.1046, + "step": 63 + }, + { + "epoch": 0.63, + "learning_rate": 5.311778173531847e-05, + "loss": 2.0074, + "step": 64 + }, + { + "epoch": 0.64, + "learning_rate": 5.2703968634527514e-05, + "loss": 2.1667, + "step": 65 + }, + { + "epoch": 0.65, + "learning_rate": 5.2284746442783414e-05, + "loss": 2.0677, + "step": 66 + }, + { + "epoch": 0.66, + "learning_rate": 5.186022739600408e-05, + "loss": 2.1798, + "step": 67 + }, + { + "epoch": 0.67, + "learning_rate": 5.143052514820357e-05, + "loss": 2.105, + "step": 68 + }, + { + "epoch": 0.68, + "learning_rate": 5.099575474106419e-05, + "loss": 2.118, + "step": 69 + }, + { + "epoch": 0.69, + "learning_rate": 5.055603257313707e-05, + "loss": 2.0791, + "step": 70 + }, + { + "epoch": 0.7, + "learning_rate": 5.011147636867943e-05, + "loss": 2.032, + "step": 71 + }, + { + "epoch": 0.71, + "learning_rate": 4.9662205146136955e-05, + "loss": 1.9331, + "step": 72 + }, + { + "epoch": 0.72, + "learning_rate": 4.92083391862797e-05, + "loss": 2.0077, + "step": 73 + }, + { + "epoch": 0.73, + "learning_rate": 4.874999999999999e-05, + "loss": 2.0253, + "step": 74 + }, + { + "epoch": 0.74, + "learning_rate": 4.828731029578105e-05, + "loss": 1.9922, + "step": 75 + }, + { + "epoch": 0.75, + "learning_rate": 4.7820393946844926e-05, + "loss": 2.0295, + "step": 76 + }, + { + "epoch": 0.76, + "learning_rate": 4.734937595798867e-05, + "loss": 1.9514, + "step": 77 + }, + { + "epoch": 0.77, + "learning_rate": 4.687438243211754e-05, + "loss": 2.1018, + "step": 78 + }, + { + "epoch": 0.78, + "learning_rate": 4.639554053648416e-05, + "loss": 2.1032, + "step": 79 + }, + { + "epoch": 0.79, + "learning_rate": 4.5912978468642824e-05, + "loss": 2.1258, + "step": 80 + }, + { + "epoch": 0.79, + "eval_loss": 1.972991704940796, + "eval_runtime": 1.494, + "eval_samples_per_second": 5.355, + "eval_steps_per_second": 0.669, + "step": 80 + }, + { + "epoch": 0.8, + "learning_rate": 4.542682542212785e-05, + "loss": 2.0353, + "step": 81 + }, + { + "epoch": 0.81, + "learning_rate": 4.4937211551865415e-05, + "loss": 1.9987, + "step": 82 + }, + { + "epoch": 0.82, + "learning_rate": 4.444426793932787e-05, + "loss": 1.8856, + "step": 83 + }, + { + "epoch": 0.83, + "learning_rate": 4.3948126557440085e-05, + "loss": 2.0904, + "step": 84 + }, + { + "epoch": 0.84, + "learning_rate": 4.344892023524714e-05, + "loss": 2.0685, + "step": 85 + }, + { + "epoch": 0.85, + "learning_rate": 4.2946782622352746e-05, + "loss": 2.1, + "step": 86 + }, + { + "epoch": 0.86, + "learning_rate": 4.244184815313799e-05, + "loss": 1.9322, + "step": 87 + }, + { + "epoch": 0.87, + "learning_rate": 4.193425201077002e-05, + "loss": 2.194, + "step": 88 + }, + { + "epoch": 0.88, + "learning_rate": 4.14241300910103e-05, + "loss": 2.0455, + "step": 89 + }, + { + "epoch": 0.89, + "learning_rate": 4.091161896583192e-05, + "loss": 2.1239, + "step": 90 + }, + { + "epoch": 0.9, + "learning_rate": 4.0396855846856076e-05, + "loss": 2.142, + "step": 91 + }, + { + "epoch": 0.91, + "learning_rate": 3.9879978548617125e-05, + "loss": 2.2304, + "step": 92 + }, + { + "epoch": 0.92, + "learning_rate": 3.936112545166636e-05, + "loss": 1.925, + "step": 93 + }, + { + "epoch": 0.93, + "learning_rate": 3.884043546552417e-05, + "loss": 2.1547, + "step": 94 + }, + { + "epoch": 0.94, + "learning_rate": 3.831804799149057e-05, + "loss": 2.0439, + "step": 95 + }, + { + "epoch": 0.95, + "learning_rate": 3.779410288532413e-05, + "loss": 1.9763, + "step": 96 + }, + { + "epoch": 0.96, + "learning_rate": 3.726874041979925e-05, + "loss": 2.114, + "step": 97 + }, + { + "epoch": 0.97, + "learning_rate": 3.674210124715168e-05, + "loss": 2.0293, + "step": 98 + }, + { + "epoch": 0.98, + "learning_rate": 3.621432636142251e-05, + "loss": 2.0833, + "step": 99 + }, + { + "epoch": 0.99, + "learning_rate": 3.5685557060710726e-05, + "loss": 2.0452, + "step": 100 + }, + { + "epoch": 0.99, + "eval_loss": 1.9682413339614868, + "eval_runtime": 1.4934, + "eval_samples_per_second": 5.357, + "eval_steps_per_second": 0.67, + "step": 100 + }, + { + "epoch": 1.0, + "learning_rate": 3.5155934909344214e-05, + "loss": 1.9682, + "step": 101 + }, + { + "epoch": 1.01, + "learning_rate": 3.462560169997965e-05, + "loss": 2.0723, + "step": 102 + }, + { + "epoch": 1.02, + "learning_rate": 3.409469941564109e-05, + "loss": 2.0713, + "step": 103 + }, + { + "epoch": 1.03, + "learning_rate": 3.356337019170772e-05, + "loss": 2.067, + "step": 104 + }, + { + "epoch": 1.04, + "learning_rate": 3.303175627786082e-05, + "loss": 1.934, + "step": 105 + }, + { + "epoch": 1.05, + "learning_rate": 3.25e-05, + "loss": 2.2083, + "step": 106 + }, + { + "epoch": 1.06, + "learning_rate": 3.1968243722139173e-05, + "loss": 2.045, + "step": 107 + }, + { + "epoch": 1.07, + "learning_rate": 3.143662980829228e-05, + "loss": 1.9531, + "step": 108 + }, + { + "epoch": 1.08, + "learning_rate": 3.090530058435891e-05, + "loss": 2.0913, + "step": 109 + }, + { + "epoch": 1.09, + "learning_rate": 3.0374398300020347e-05, + "loss": 1.9211, + "step": 110 + }, + { + "epoch": 1.1, + "learning_rate": 2.984406509065579e-05, + "loss": 2.0983, + "step": 111 + }, + { + "epoch": 1.11, + "learning_rate": 2.9314442939289275e-05, + "loss": 2.1621, + "step": 112 + }, + { + "epoch": 1.12, + "learning_rate": 2.8785673638577486e-05, + "loss": 2.1728, + "step": 113 + }, + { + "epoch": 1.13, + "learning_rate": 2.825789875284833e-05, + "loss": 2.0512, + "step": 114 + }, + { + "epoch": 1.14, + "learning_rate": 2.773125958020074e-05, + "loss": 2.0462, + "step": 115 + }, + { + "epoch": 1.15, + "learning_rate": 2.7205897114675866e-05, + "loss": 1.9357, + "step": 116 + }, + { + "epoch": 1.16, + "learning_rate": 2.668195200850944e-05, + "loss": 1.8384, + "step": 117 + }, + { + "epoch": 1.17, + "learning_rate": 2.6159564534475832e-05, + "loss": 1.9649, + "step": 118 + }, + { + "epoch": 1.18, + "learning_rate": 2.563887454833363e-05, + "loss": 2.0262, + "step": 119 + }, + { + "epoch": 1.19, + "learning_rate": 2.5120021451382872e-05, + "loss": 2.1315, + "step": 120 + }, + { + "epoch": 1.19, + "eval_loss": 1.9639294147491455, + "eval_runtime": 1.4959, + "eval_samples_per_second": 5.348, + "eval_steps_per_second": 0.668, + "step": 120 + }, + { + "epoch": 1.2, + "learning_rate": 2.4603144153143925e-05, + "loss": 1.9284, + "step": 121 + }, + { + "epoch": 1.21, + "learning_rate": 2.4088381034168072e-05, + "loss": 2.0671, + "step": 122 + }, + { + "epoch": 1.22, + "learning_rate": 2.3575869908989705e-05, + "loss": 2.0938, + "step": 123 + }, + { + "epoch": 1.23, + "learning_rate": 2.306574798922998e-05, + "loss": 1.9189, + "step": 124 + }, + { + "epoch": 1.24, + "learning_rate": 2.2558151846862005e-05, + "loss": 2.0635, + "step": 125 + }, + { + "epoch": 1.25, + "learning_rate": 2.2053217377647255e-05, + "loss": 2.1367, + "step": 126 + }, + { + "epoch": 1.26, + "learning_rate": 2.1551079764752848e-05, + "loss": 1.9967, + "step": 127 + }, + { + "epoch": 1.27, + "learning_rate": 2.105187344255991e-05, + "loss": 2.1994, + "step": 128 + }, + { + "epoch": 1.28, + "learning_rate": 2.0555732060672138e-05, + "loss": 2.151, + "step": 129 + }, + { + "epoch": 1.29, + "learning_rate": 2.0062788448134583e-05, + "loss": 2.0517, + "step": 130 + }, + { + "epoch": 1.3, + "learning_rate": 1.957317457787214e-05, + "loss": 2.0584, + "step": 131 + }, + { + "epoch": 1.31, + "learning_rate": 1.908702153135717e-05, + "loss": 2.0868, + "step": 132 + }, + { + "epoch": 1.32, + "learning_rate": 1.860445946351584e-05, + "loss": 1.9646, + "step": 133 + }, + { + "epoch": 1.33, + "learning_rate": 1.8125617567882463e-05, + "loss": 1.986, + "step": 134 + }, + { + "epoch": 1.34, + "learning_rate": 1.7650624042011325e-05, + "loss": 1.9015, + "step": 135 + }, + { + "epoch": 1.35, + "learning_rate": 1.7179606053155072e-05, + "loss": 2.1968, + "step": 136 + }, + { + "epoch": 1.36, + "learning_rate": 1.6712689704218944e-05, + "loss": 1.9268, + "step": 137 + }, + { + "epoch": 1.37, + "learning_rate": 1.6250000000000005e-05, + "loss": 2.0205, + "step": 138 + }, + { + "epoch": 1.38, + "learning_rate": 1.5791660813720294e-05, + "loss": 2.1183, + "step": 139 + }, + { + "epoch": 1.39, + "learning_rate": 1.533779485386304e-05, + "loss": 2.08, + "step": 140 + }, + { + "epoch": 1.39, + "eval_loss": 1.9603089094161987, + "eval_runtime": 1.495, + "eval_samples_per_second": 5.351, + "eval_steps_per_second": 0.669, + "step": 140 + }, + { + "epoch": 1.4, + "learning_rate": 1.4888523631320579e-05, + "loss": 1.9309, + "step": 141 + }, + { + "epoch": 1.41, + "learning_rate": 1.4443967426862935e-05, + "loss": 2.0579, + "step": 142 + }, + { + "epoch": 1.42, + "learning_rate": 1.4004245258935799e-05, + "loss": 1.8377, + "step": 143 + }, + { + "epoch": 1.43, + "learning_rate": 1.3569474851796432e-05, + "loss": 2.0764, + "step": 144 + }, + { + "epoch": 1.44, + "learning_rate": 1.3139772603995914e-05, + "loss": 2.1289, + "step": 145 + }, + { + "epoch": 1.45, + "learning_rate": 1.2715253557216577e-05, + "loss": 2.041, + "step": 146 + }, + { + "epoch": 1.46, + "learning_rate": 1.2296031365472491e-05, + "loss": 1.9343, + "step": 147 + }, + { + "epoch": 1.47, + "learning_rate": 1.1882218264681525e-05, + "loss": 1.9665, + "step": 148 + }, + { + "epoch": 1.48, + "learning_rate": 1.1473925042617137e-05, + "loss": 2.1843, + "step": 149 + }, + { + "epoch": 1.49, + "learning_rate": 1.1071261009247762e-05, + "loss": 2.0125, + "step": 150 + } + ], + "logging_steps": 1, + "max_steps": 202, + "num_train_epochs": 2, + "save_steps": 50, + "total_flos": 1.2497211511065805e+17, + "trial_name": null, + "trial_params": null +} diff --git a/multi-prompt/twochar/checkpoint-150/training_args.bin b/multi-prompt/twochar/checkpoint-150/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..47d6200b8d4047ff364e6b15cf014dde61f2166b --- /dev/null +++ b/multi-prompt/twochar/checkpoint-150/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac3fb199292750362f1e387a5d8bde7088e81478bbe173194e512403378f1c70 +size 4475 diff --git a/multi-prompt/twochar/checkpoint-200/README.md b/multi-prompt/twochar/checkpoint-200/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/multi-prompt/twochar/checkpoint-200/adapter_config.json b/multi-prompt/twochar/checkpoint-200/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b33f8685004fc9f9eef2843be41eb6081508fc38 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.01, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "v_proj", + "q_proj", + "down_proj", + "k_proj", + "gate_proj", + "o_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/multi-prompt/twochar/checkpoint-200/adapter_model.bin b/multi-prompt/twochar/checkpoint-200/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..8087b4c30f7a0539b15177120ff755b578eba028 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c882f2a426060ad44b5cf280eb92814d684c2673107ca7f19c9e34b4c3fe2980 +size 125374989 diff --git a/multi-prompt/twochar/checkpoint-200/adapter_model/README.md b/multi-prompt/twochar/checkpoint-200/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/adapter_model/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/multi-prompt/twochar/checkpoint-200/adapter_model/adapter_config.json b/multi-prompt/twochar/checkpoint-200/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b33f8685004fc9f9eef2843be41eb6081508fc38 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/adapter_model/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.01, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "v_proj", + "q_proj", + "down_proj", + "k_proj", + "gate_proj", + "o_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/multi-prompt/twochar/checkpoint-200/adapter_model/adapter_model.bin b/multi-prompt/twochar/checkpoint-200/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..8087b4c30f7a0539b15177120ff755b578eba028 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c882f2a426060ad44b5cf280eb92814d684c2673107ca7f19c9e34b4c3fe2980 +size 125374989 diff --git a/multi-prompt/twochar/checkpoint-200/optimizer.pt b/multi-prompt/twochar/checkpoint-200/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..3571f4113bbbf3511ef6bbc74d021d6e95cb81f9 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:16b546d91cb67c74c723898ad8350b04c2ee27be0612fff40324b5db78dd87a0 +size 250681597 diff --git a/multi-prompt/twochar/checkpoint-200/rng_state_0.pth b/multi-prompt/twochar/checkpoint-200/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..f7672dac941e999991ba5ab5e715528b64140eb3 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:19c4ecd007574b31ef585631572d90a849b7178b232262c2c41125e361bc3517 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-200/rng_state_1.pth b/multi-prompt/twochar/checkpoint-200/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..f026ecd896be0127311bc707250b1058306d0ff9 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3cc05c6dead1d35e2b0a4b8c8ca26e5b7aeb7547f432cc1c89591900a46a09b1 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-200/rng_state_2.pth b/multi-prompt/twochar/checkpoint-200/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..85110d5a642c4bd071c7683f564e33f3a00ef14f --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:10c2d8765af7ee78c7ddacf8d9c6ad56309da3711e52aa502ca9e611e8b23df8 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-200/rng_state_3.pth b/multi-prompt/twochar/checkpoint-200/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..86e83fa6afa0df4b6812870b44ce98497c19bcdc --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f581ef0ddd2f517335879867250c3475ad9cf8ee68bb06bf8b999b0ce994eaff +size 21687 diff --git a/multi-prompt/twochar/checkpoint-200/rng_state_4.pth b/multi-prompt/twochar/checkpoint-200/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..ea54c5d05db733142fd6a2b82a00f09b4f9c4962 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:79d2080ee3cbb2aa61b39ad477ae141bb73b045d6857eaac0076468a913cf2e7 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-200/rng_state_5.pth b/multi-prompt/twochar/checkpoint-200/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..ccb27daceb7ac2a3543fd1c8ca4a8a6745d4f2ab --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8436025ee61c8b1e2881effd13c23c89bf56b5d39765503ff7e4d1ff287fb833 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-200/rng_state_6.pth b/multi-prompt/twochar/checkpoint-200/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..10bdadd541a4844d4f34266c66033c8426994fe4 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2e172e6c55f1ad800d794787541c81bf86fbab4bc0b983c42152c182f34992c0 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-200/rng_state_7.pth b/multi-prompt/twochar/checkpoint-200/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..e088b0c18a3ce89c1ec7c68ecc0ba07e135370e1 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9547c06d3d25b5745d0cbb3abc1cd35c3aa5dcebb97f2bb750d73d134c8e4a73 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-200/scheduler.pt b/multi-prompt/twochar/checkpoint-200/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..e663fbc8d5230eaa608d8784982b5358e06d4d6f --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb782d53827be80b0bd501ee71f6a9c209f4fa9b3850b8515f3c0b99d0f7e4f7 +size 627 diff --git a/multi-prompt/twochar/checkpoint-200/trainer_state.json b/multi-prompt/twochar/checkpoint-200/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..56075ef7c10a941c1d8ddfd8b4cf58f56a2b6d9e --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/trainer_state.json @@ -0,0 +1,1299 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.9801980198019802, + "eval_steps": 20, + "global_step": 200, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.01, + "learning_rate": 6.5e-06, + "loss": 2.0407, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 1.3e-05, + "loss": 2.3107, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 1.9499999999999996e-05, + "loss": 2.1613, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 2.6e-05, + "loss": 2.0444, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 3.25e-05, + "loss": 2.2512, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 3.899999999999999e-05, + "loss": 2.0516, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 4.5499999999999995e-05, + "loss": 2.2006, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 5.2e-05, + "loss": 2.1229, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 5.85e-05, + "loss": 2.2498, + "step": 9 + }, + { + "epoch": 0.1, + "learning_rate": 6.5e-05, + "loss": 2.0293, + "step": 10 + }, + { + "epoch": 0.11, + "learning_rate": 6.499564948206075e-05, + "loss": 2.1316, + "step": 11 + }, + { + "epoch": 0.12, + "learning_rate": 6.498259909298188e-05, + "loss": 2.1274, + "step": 12 + }, + { + "epoch": 0.13, + "learning_rate": 6.49608523266681e-05, + "loss": 2.1129, + "step": 13 + }, + { + "epoch": 0.14, + "learning_rate": 6.493041500525461e-05, + "loss": 2.1517, + "step": 14 + }, + { + "epoch": 0.15, + "learning_rate": 6.489129527754836e-05, + "loss": 2.1078, + "step": 15 + }, + { + "epoch": 0.16, + "learning_rate": 6.484350361684639e-05, + "loss": 2.0362, + "step": 16 + }, + { + "epoch": 0.17, + "learning_rate": 6.478705281813194e-05, + "loss": 2.0553, + "step": 17 + }, + { + "epoch": 0.18, + "learning_rate": 6.472195799464884e-05, + "loss": 2.1223, + "step": 18 + }, + { + "epoch": 0.19, + "learning_rate": 6.464823657385538e-05, + "loss": 2.0326, + "step": 19 + }, + { + "epoch": 0.2, + "learning_rate": 6.456590829275857e-05, + "loss": 2.1802, + "step": 20 + }, + { + "epoch": 0.2, + "eval_loss": 2.012301445007324, + "eval_runtime": 1.4855, + "eval_samples_per_second": 5.385, + "eval_steps_per_second": 0.673, + "step": 20 + }, + { + "epoch": 0.21, + "learning_rate": 6.447499519263001e-05, + "loss": 2.0665, + "step": 21 + }, + { + "epoch": 0.22, + "learning_rate": 6.437552161310498e-05, + "loss": 2.0354, + "step": 22 + }, + { + "epoch": 0.23, + "learning_rate": 6.426751418566609e-05, + "loss": 2.1289, + "step": 23 + }, + { + "epoch": 0.24, + "learning_rate": 6.415100182651334e-05, + "loss": 2.0742, + "step": 24 + }, + { + "epoch": 0.25, + "learning_rate": 6.402601572882268e-05, + "loss": 2.0412, + "step": 25 + }, + { + "epoch": 0.26, + "learning_rate": 6.38925893543947e-05, + "loss": 2.1974, + "step": 26 + }, + { + "epoch": 0.27, + "learning_rate": 6.375075842469626e-05, + "loss": 2.196, + "step": 27 + }, + { + "epoch": 0.28, + "learning_rate": 6.360056091129678e-05, + "loss": 2.0775, + "step": 28 + }, + { + "epoch": 0.29, + "learning_rate": 6.344203702570254e-05, + "loss": 2.0366, + "step": 29 + }, + { + "epoch": 0.3, + "learning_rate": 6.327522920859093e-05, + "loss": 2.1438, + "step": 30 + }, + { + "epoch": 0.31, + "learning_rate": 6.310018211844817e-05, + "loss": 2.0639, + "step": 31 + }, + { + "epoch": 0.32, + "learning_rate": 6.291694261961308e-05, + "loss": 2.1794, + "step": 32 + }, + { + "epoch": 0.33, + "learning_rate": 6.272555976973039e-05, + "loss": 2.0592, + "step": 33 + }, + { + "epoch": 0.34, + "learning_rate": 6.252608480661681e-05, + "loss": 2.1628, + "step": 34 + }, + { + "epoch": 0.35, + "learning_rate": 6.231857113454346e-05, + "loss": 2.1851, + "step": 35 + }, + { + "epoch": 0.36, + "learning_rate": 6.21030743099382e-05, + "loss": 2.0531, + "step": 36 + }, + { + "epoch": 0.37, + "learning_rate": 6.18796520265119e-05, + "loss": 2.0284, + "step": 37 + }, + { + "epoch": 0.38, + "learning_rate": 6.164836409981237e-05, + "loss": 2.2359, + "step": 38 + }, + { + "epoch": 0.39, + "learning_rate": 6.140927245121032e-05, + "loss": 1.8967, + "step": 39 + }, + { + "epoch": 0.4, + "learning_rate": 6.116244109132153e-05, + "loss": 2.0275, + "step": 40 + }, + { + "epoch": 0.4, + "eval_loss": 1.9861761331558228, + "eval_runtime": 1.4896, + "eval_samples_per_second": 5.37, + "eval_steps_per_second": 0.671, + "step": 40 + }, + { + "epoch": 0.41, + "learning_rate": 6.0907936102869656e-05, + "loss": 2.0316, + "step": 41 + }, + { + "epoch": 0.42, + "learning_rate": 6.0645825622994254e-05, + "loss": 2.0768, + "step": 42 + }, + { + "epoch": 0.43, + "learning_rate": 6.037617982500884e-05, + "loss": 2.0058, + "step": 43 + }, + { + "epoch": 0.44, + "learning_rate": 6.009907089961381e-05, + "loss": 2.2319, + "step": 44 + }, + { + "epoch": 0.45, + "learning_rate": 5.981457303556923e-05, + "loss": 2.1693, + "step": 45 + }, + { + "epoch": 0.46, + "learning_rate": 5.9522762399832716e-05, + "loss": 2.0858, + "step": 46 + }, + { + "epoch": 0.47, + "learning_rate": 5.922371711716768e-05, + "loss": 2.0576, + "step": 47 + }, + { + "epoch": 0.48, + "learning_rate": 5.891751724922749e-05, + "loss": 1.9946, + "step": 48 + }, + { + "epoch": 0.49, + "learning_rate": 5.860424477312095e-05, + "loss": 2.0166, + "step": 49 + }, + { + "epoch": 0.5, + "learning_rate": 5.828398355946514e-05, + "loss": 2.1482, + "step": 50 + }, + { + "epoch": 0.5, + "learning_rate": 5.795681934993113e-05, + "loss": 2.0698, + "step": 51 + }, + { + "epoch": 0.51, + "learning_rate": 5.7622839734288945e-05, + "loss": 1.9317, + "step": 52 + }, + { + "epoch": 0.52, + "learning_rate": 5.728213412695761e-05, + "loss": 2.1399, + "step": 53 + }, + { + "epoch": 0.53, + "learning_rate": 5.693479374306676e-05, + "loss": 2.1637, + "step": 54 + }, + { + "epoch": 0.54, + "learning_rate": 5.658091157403617e-05, + "loss": 2.0557, + "step": 55 + }, + { + "epoch": 0.55, + "learning_rate": 5.622058236267965e-05, + "loss": 1.926, + "step": 56 + }, + { + "epoch": 0.56, + "learning_rate": 5.585390257784018e-05, + "loss": 2.0984, + "step": 57 + }, + { + "epoch": 0.57, + "learning_rate": 5.548097038856279e-05, + "loss": 2.0428, + "step": 58 + }, + { + "epoch": 0.58, + "learning_rate": 5.5101885637812374e-05, + "loss": 1.9624, + "step": 59 + }, + { + "epoch": 0.59, + "learning_rate": 5.4716749815743304e-05, + "loss": 1.913, + "step": 60 + }, + { + "epoch": 0.59, + "eval_loss": 1.97765052318573, + "eval_runtime": 1.4886, + "eval_samples_per_second": 5.374, + "eval_steps_per_second": 0.672, + "step": 60 + }, + { + "epoch": 0.6, + "learning_rate": 5.432566603252809e-05, + "loss": 2.0125, + "step": 61 + }, + { + "epoch": 0.61, + "learning_rate": 5.3928738990752234e-05, + "loss": 2.0721, + "step": 62 + }, + { + "epoch": 0.62, + "learning_rate": 5.3526074957382866e-05, + "loss": 2.1046, + "step": 63 + }, + { + "epoch": 0.63, + "learning_rate": 5.311778173531847e-05, + "loss": 2.0074, + "step": 64 + }, + { + "epoch": 0.64, + "learning_rate": 5.2703968634527514e-05, + "loss": 2.1667, + "step": 65 + }, + { + "epoch": 0.65, + "learning_rate": 5.2284746442783414e-05, + "loss": 2.0677, + "step": 66 + }, + { + "epoch": 0.66, + "learning_rate": 5.186022739600408e-05, + "loss": 2.1798, + "step": 67 + }, + { + "epoch": 0.67, + "learning_rate": 5.143052514820357e-05, + "loss": 2.105, + "step": 68 + }, + { + "epoch": 0.68, + "learning_rate": 5.099575474106419e-05, + "loss": 2.118, + "step": 69 + }, + { + "epoch": 0.69, + "learning_rate": 5.055603257313707e-05, + "loss": 2.0791, + "step": 70 + }, + { + "epoch": 0.7, + "learning_rate": 5.011147636867943e-05, + "loss": 2.032, + "step": 71 + }, + { + "epoch": 0.71, + "learning_rate": 4.9662205146136955e-05, + "loss": 1.9331, + "step": 72 + }, + { + "epoch": 0.72, + "learning_rate": 4.92083391862797e-05, + "loss": 2.0077, + "step": 73 + }, + { + "epoch": 0.73, + "learning_rate": 4.874999999999999e-05, + "loss": 2.0253, + "step": 74 + }, + { + "epoch": 0.74, + "learning_rate": 4.828731029578105e-05, + "loss": 1.9922, + "step": 75 + }, + { + "epoch": 0.75, + "learning_rate": 4.7820393946844926e-05, + "loss": 2.0295, + "step": 76 + }, + { + "epoch": 0.76, + "learning_rate": 4.734937595798867e-05, + "loss": 1.9514, + "step": 77 + }, + { + "epoch": 0.77, + "learning_rate": 4.687438243211754e-05, + "loss": 2.1018, + "step": 78 + }, + { + "epoch": 0.78, + "learning_rate": 4.639554053648416e-05, + "loss": 2.1032, + "step": 79 + }, + { + "epoch": 0.79, + "learning_rate": 4.5912978468642824e-05, + "loss": 2.1258, + "step": 80 + }, + { + "epoch": 0.79, + "eval_loss": 1.972991704940796, + "eval_runtime": 1.494, + "eval_samples_per_second": 5.355, + "eval_steps_per_second": 0.669, + "step": 80 + }, + { + "epoch": 0.8, + "learning_rate": 4.542682542212785e-05, + "loss": 2.0353, + "step": 81 + }, + { + "epoch": 0.81, + "learning_rate": 4.4937211551865415e-05, + "loss": 1.9987, + "step": 82 + }, + { + "epoch": 0.82, + "learning_rate": 4.444426793932787e-05, + "loss": 1.8856, + "step": 83 + }, + { + "epoch": 0.83, + "learning_rate": 4.3948126557440085e-05, + "loss": 2.0904, + "step": 84 + }, + { + "epoch": 0.84, + "learning_rate": 4.344892023524714e-05, + "loss": 2.0685, + "step": 85 + }, + { + "epoch": 0.85, + "learning_rate": 4.2946782622352746e-05, + "loss": 2.1, + "step": 86 + }, + { + "epoch": 0.86, + "learning_rate": 4.244184815313799e-05, + "loss": 1.9322, + "step": 87 + }, + { + "epoch": 0.87, + "learning_rate": 4.193425201077002e-05, + "loss": 2.194, + "step": 88 + }, + { + "epoch": 0.88, + "learning_rate": 4.14241300910103e-05, + "loss": 2.0455, + "step": 89 + }, + { + "epoch": 0.89, + "learning_rate": 4.091161896583192e-05, + "loss": 2.1239, + "step": 90 + }, + { + "epoch": 0.9, + "learning_rate": 4.0396855846856076e-05, + "loss": 2.142, + "step": 91 + }, + { + "epoch": 0.91, + "learning_rate": 3.9879978548617125e-05, + "loss": 2.2304, + "step": 92 + }, + { + "epoch": 0.92, + "learning_rate": 3.936112545166636e-05, + "loss": 1.925, + "step": 93 + }, + { + "epoch": 0.93, + "learning_rate": 3.884043546552417e-05, + "loss": 2.1547, + "step": 94 + }, + { + "epoch": 0.94, + "learning_rate": 3.831804799149057e-05, + "loss": 2.0439, + "step": 95 + }, + { + "epoch": 0.95, + "learning_rate": 3.779410288532413e-05, + "loss": 1.9763, + "step": 96 + }, + { + "epoch": 0.96, + "learning_rate": 3.726874041979925e-05, + "loss": 2.114, + "step": 97 + }, + { + "epoch": 0.97, + "learning_rate": 3.674210124715168e-05, + "loss": 2.0293, + "step": 98 + }, + { + "epoch": 0.98, + "learning_rate": 3.621432636142251e-05, + "loss": 2.0833, + "step": 99 + }, + { + "epoch": 0.99, + "learning_rate": 3.5685557060710726e-05, + "loss": 2.0452, + "step": 100 + }, + { + "epoch": 0.99, + "eval_loss": 1.9682413339614868, + "eval_runtime": 1.4934, + "eval_samples_per_second": 5.357, + "eval_steps_per_second": 0.67, + "step": 100 + }, + { + "epoch": 1.0, + "learning_rate": 3.5155934909344214e-05, + "loss": 1.9682, + "step": 101 + }, + { + "epoch": 1.01, + "learning_rate": 3.462560169997965e-05, + "loss": 2.0723, + "step": 102 + }, + { + "epoch": 1.02, + "learning_rate": 3.409469941564109e-05, + "loss": 2.0713, + "step": 103 + }, + { + "epoch": 1.03, + "learning_rate": 3.356337019170772e-05, + "loss": 2.067, + "step": 104 + }, + { + "epoch": 1.04, + "learning_rate": 3.303175627786082e-05, + "loss": 1.934, + "step": 105 + }, + { + "epoch": 1.05, + "learning_rate": 3.25e-05, + "loss": 2.2083, + "step": 106 + }, + { + "epoch": 1.06, + "learning_rate": 3.1968243722139173e-05, + "loss": 2.045, + "step": 107 + }, + { + "epoch": 1.07, + "learning_rate": 3.143662980829228e-05, + "loss": 1.9531, + "step": 108 + }, + { + "epoch": 1.08, + "learning_rate": 3.090530058435891e-05, + "loss": 2.0913, + "step": 109 + }, + { + "epoch": 1.09, + "learning_rate": 3.0374398300020347e-05, + "loss": 1.9211, + "step": 110 + }, + { + "epoch": 1.1, + "learning_rate": 2.984406509065579e-05, + "loss": 2.0983, + "step": 111 + }, + { + "epoch": 1.11, + "learning_rate": 2.9314442939289275e-05, + "loss": 2.1621, + "step": 112 + }, + { + "epoch": 1.12, + "learning_rate": 2.8785673638577486e-05, + "loss": 2.1728, + "step": 113 + }, + { + "epoch": 1.13, + "learning_rate": 2.825789875284833e-05, + "loss": 2.0512, + "step": 114 + }, + { + "epoch": 1.14, + "learning_rate": 2.773125958020074e-05, + "loss": 2.0462, + "step": 115 + }, + { + "epoch": 1.15, + "learning_rate": 2.7205897114675866e-05, + "loss": 1.9357, + "step": 116 + }, + { + "epoch": 1.16, + "learning_rate": 2.668195200850944e-05, + "loss": 1.8384, + "step": 117 + }, + { + "epoch": 1.17, + "learning_rate": 2.6159564534475832e-05, + "loss": 1.9649, + "step": 118 + }, + { + "epoch": 1.18, + "learning_rate": 2.563887454833363e-05, + "loss": 2.0262, + "step": 119 + }, + { + "epoch": 1.19, + "learning_rate": 2.5120021451382872e-05, + "loss": 2.1315, + "step": 120 + }, + { + "epoch": 1.19, + "eval_loss": 1.9639294147491455, + "eval_runtime": 1.4959, + "eval_samples_per_second": 5.348, + "eval_steps_per_second": 0.668, + "step": 120 + }, + { + "epoch": 1.2, + "learning_rate": 2.4603144153143925e-05, + "loss": 1.9284, + "step": 121 + }, + { + "epoch": 1.21, + "learning_rate": 2.4088381034168072e-05, + "loss": 2.0671, + "step": 122 + }, + { + "epoch": 1.22, + "learning_rate": 2.3575869908989705e-05, + "loss": 2.0938, + "step": 123 + }, + { + "epoch": 1.23, + "learning_rate": 2.306574798922998e-05, + "loss": 1.9189, + "step": 124 + }, + { + "epoch": 1.24, + "learning_rate": 2.2558151846862005e-05, + "loss": 2.0635, + "step": 125 + }, + { + "epoch": 1.25, + "learning_rate": 2.2053217377647255e-05, + "loss": 2.1367, + "step": 126 + }, + { + "epoch": 1.26, + "learning_rate": 2.1551079764752848e-05, + "loss": 1.9967, + "step": 127 + }, + { + "epoch": 1.27, + "learning_rate": 2.105187344255991e-05, + "loss": 2.1994, + "step": 128 + }, + { + "epoch": 1.28, + "learning_rate": 2.0555732060672138e-05, + "loss": 2.151, + "step": 129 + }, + { + "epoch": 1.29, + "learning_rate": 2.0062788448134583e-05, + "loss": 2.0517, + "step": 130 + }, + { + "epoch": 1.3, + "learning_rate": 1.957317457787214e-05, + "loss": 2.0584, + "step": 131 + }, + { + "epoch": 1.31, + "learning_rate": 1.908702153135717e-05, + "loss": 2.0868, + "step": 132 + }, + { + "epoch": 1.32, + "learning_rate": 1.860445946351584e-05, + "loss": 1.9646, + "step": 133 + }, + { + "epoch": 1.33, + "learning_rate": 1.8125617567882463e-05, + "loss": 1.986, + "step": 134 + }, + { + "epoch": 1.34, + "learning_rate": 1.7650624042011325e-05, + "loss": 1.9015, + "step": 135 + }, + { + "epoch": 1.35, + "learning_rate": 1.7179606053155072e-05, + "loss": 2.1968, + "step": 136 + }, + { + "epoch": 1.36, + "learning_rate": 1.6712689704218944e-05, + "loss": 1.9268, + "step": 137 + }, + { + "epoch": 1.37, + "learning_rate": 1.6250000000000005e-05, + "loss": 2.0205, + "step": 138 + }, + { + "epoch": 1.38, + "learning_rate": 1.5791660813720294e-05, + "loss": 2.1183, + "step": 139 + }, + { + "epoch": 1.39, + "learning_rate": 1.533779485386304e-05, + "loss": 2.08, + "step": 140 + }, + { + "epoch": 1.39, + "eval_loss": 1.9603089094161987, + "eval_runtime": 1.495, + "eval_samples_per_second": 5.351, + "eval_steps_per_second": 0.669, + "step": 140 + }, + { + "epoch": 1.4, + "learning_rate": 1.4888523631320579e-05, + "loss": 1.9309, + "step": 141 + }, + { + "epoch": 1.41, + "learning_rate": 1.4443967426862935e-05, + "loss": 2.0579, + "step": 142 + }, + { + "epoch": 1.42, + "learning_rate": 1.4004245258935799e-05, + "loss": 1.8377, + "step": 143 + }, + { + "epoch": 1.43, + "learning_rate": 1.3569474851796432e-05, + "loss": 2.0764, + "step": 144 + }, + { + "epoch": 1.44, + "learning_rate": 1.3139772603995914e-05, + "loss": 2.1289, + "step": 145 + }, + { + "epoch": 1.45, + "learning_rate": 1.2715253557216577e-05, + "loss": 2.041, + "step": 146 + }, + { + "epoch": 1.46, + "learning_rate": 1.2296031365472491e-05, + "loss": 1.9343, + "step": 147 + }, + { + "epoch": 1.47, + "learning_rate": 1.1882218264681525e-05, + "loss": 1.9665, + "step": 148 + }, + { + "epoch": 1.48, + "learning_rate": 1.1473925042617137e-05, + "loss": 2.1843, + "step": 149 + }, + { + "epoch": 1.49, + "learning_rate": 1.1071261009247762e-05, + "loss": 2.0125, + "step": 150 + }, + { + "epoch": 1.5, + "learning_rate": 1.06743339674719e-05, + "loss": 2.1792, + "step": 151 + }, + { + "epoch": 1.5, + "learning_rate": 1.028325018425668e-05, + "loss": 1.9815, + "step": 152 + }, + { + "epoch": 1.51, + "learning_rate": 9.898114362187623e-06, + "loss": 1.9769, + "step": 153 + }, + { + "epoch": 1.52, + "learning_rate": 9.519029611437207e-06, + "loss": 1.9301, + "step": 154 + }, + { + "epoch": 1.53, + "learning_rate": 9.146097422159812e-06, + "loss": 1.979, + "step": 155 + }, + { + "epoch": 1.54, + "learning_rate": 8.77941763732035e-06, + "loss": 2.0074, + "step": 156 + }, + { + "epoch": 1.55, + "learning_rate": 8.419088425963836e-06, + "loss": 2.2051, + "step": 157 + }, + { + "epoch": 1.56, + "learning_rate": 8.065206256933238e-06, + "loss": 2.0702, + "step": 158 + }, + { + "epoch": 1.57, + "learning_rate": 7.717865873042387e-06, + "loss": 1.9569, + "step": 159 + }, + { + "epoch": 1.58, + "learning_rate": 7.377160265711047e-06, + "loss": 2.0747, + "step": 160 + }, + { + "epoch": 1.58, + "eval_loss": 1.9598100185394287, + "eval_runtime": 1.5252, + "eval_samples_per_second": 5.245, + "eval_steps_per_second": 0.656, + "step": 160 + }, + { + "epoch": 1.59, + "learning_rate": 7.043180650068862e-06, + "loss": 2.0023, + "step": 161 + }, + { + "epoch": 1.6, + "learning_rate": 6.71601644053486e-06, + "loss": 1.8974, + "step": 162 + }, + { + "epoch": 1.61, + "learning_rate": 6.395755226879042e-06, + "loss": 2.0356, + "step": 163 + }, + { + "epoch": 1.62, + "learning_rate": 6.082482750772508e-06, + "loss": 2.0573, + "step": 164 + }, + { + "epoch": 1.63, + "learning_rate": 5.776282882832316e-06, + "loss": 2.188, + "step": 165 + }, + { + "epoch": 1.64, + "learning_rate": 5.477237600167276e-06, + "loss": 1.9582, + "step": 166 + }, + { + "epoch": 1.65, + "learning_rate": 5.185426964430758e-06, + "loss": 1.9744, + "step": 167 + }, + { + "epoch": 1.66, + "learning_rate": 4.9009291003861895e-06, + "loss": 1.9664, + "step": 168 + }, + { + "epoch": 1.67, + "learning_rate": 4.623820174991159e-06, + "loss": 1.9808, + "step": 169 + }, + { + "epoch": 1.68, + "learning_rate": 4.3541743770057415e-06, + "loss": 2.0246, + "step": 170 + }, + { + "epoch": 1.69, + "learning_rate": 4.092063897130342e-06, + "loss": 2.0081, + "step": 171 + }, + { + "epoch": 1.7, + "learning_rate": 3.837558908678464e-06, + "loss": 2.0681, + "step": 172 + }, + { + "epoch": 1.71, + "learning_rate": 3.590727548789681e-06, + "loss": 1.9028, + "step": 173 + }, + { + "epoch": 1.72, + "learning_rate": 3.3516359001876314e-06, + "loss": 2.0582, + "step": 174 + }, + { + "epoch": 1.73, + "learning_rate": 3.120347973488091e-06, + "loss": 2.0035, + "step": 175 + }, + { + "epoch": 1.74, + "learning_rate": 2.896925690061783e-06, + "loss": 1.8968, + "step": 176 + }, + { + "epoch": 1.75, + "learning_rate": 2.6814288654565337e-06, + "loss": 2.103, + "step": 177 + }, + { + "epoch": 1.76, + "learning_rate": 2.473915193383181e-06, + "loss": 1.9192, + "step": 178 + }, + { + "epoch": 1.77, + "learning_rate": 2.2744402302696074e-06, + "loss": 2.0644, + "step": 179 + }, + { + "epoch": 1.78, + "learning_rate": 2.08305738038692e-06, + "loss": 2.0469, + "step": 180 + }, + { + "epoch": 1.78, + "eval_loss": 1.958908200263977, + "eval_runtime": 1.5285, + "eval_samples_per_second": 5.234, + "eval_steps_per_second": 0.654, + "step": 180 + }, + { + "epoch": 1.79, + "learning_rate": 1.8998178815518273e-06, + "loss": 2.0494, + "step": 181 + }, + { + "epoch": 1.8, + "learning_rate": 1.7247707914090686e-06, + "loss": 2.0655, + "step": 182 + }, + { + "epoch": 1.81, + "learning_rate": 1.557962974297461e-06, + "loss": 2.1354, + "step": 183 + }, + { + "epoch": 1.82, + "learning_rate": 1.3994390887032131e-06, + "loss": 2.1334, + "step": 184 + }, + { + "epoch": 1.83, + "learning_rate": 1.2492415753037343e-06, + "loss": 1.9995, + "step": 185 + }, + { + "epoch": 1.84, + "learning_rate": 1.1074106456052833e-06, + "loss": 2.0004, + "step": 186 + }, + { + "epoch": 1.85, + "learning_rate": 9.739842711773208e-07, + "loss": 1.9561, + "step": 187 + }, + { + "epoch": 1.86, + "learning_rate": 8.489981734866568e-07, + "loss": 2.0507, + "step": 188 + }, + { + "epoch": 1.87, + "learning_rate": 7.324858143339169e-07, + "loss": 2.0923, + "step": 189 + }, + { + "epoch": 1.88, + "learning_rate": 6.24478386895011e-07, + "loss": 1.9369, + "step": 190 + }, + { + "epoch": 1.89, + "learning_rate": 5.250048073699814e-07, + "loss": 2.0687, + "step": 191 + }, + { + "epoch": 1.9, + "learning_rate": 4.3409170724143323e-07, + "loss": 1.9242, + "step": 192 + }, + { + "epoch": 1.91, + "learning_rate": 3.51763426144617e-07, + "loss": 1.9617, + "step": 193 + }, + { + "epoch": 1.92, + "learning_rate": 2.7804200535116254e-07, + "loss": 1.9776, + "step": 194 + }, + { + "epoch": 1.93, + "learning_rate": 2.1294718186805808e-07, + "loss": 2.0486, + "step": 195 + }, + { + "epoch": 1.94, + "learning_rate": 1.564963831536034e-07, + "loss": 1.8825, + "step": 196 + }, + { + "epoch": 1.95, + "learning_rate": 1.0870472245163964e-07, + "loss": 2.0531, + "step": 197 + }, + { + "epoch": 1.96, + "learning_rate": 6.958499474538692e-08, + "loss": 2.0535, + "step": 198 + }, + { + "epoch": 1.97, + "learning_rate": 3.9147673331896836e-08, + "loss": 2.068, + "step": 199 + }, + { + "epoch": 1.98, + "learning_rate": 1.740090701811525e-08, + "loss": 2.0969, + "step": 200 + }, + { + "epoch": 1.98, + "eval_loss": 1.9594972133636475, + "eval_runtime": 1.4959, + "eval_samples_per_second": 5.348, + "eval_steps_per_second": 0.668, + "step": 200 + } + ], + "logging_steps": 1, + "max_steps": 202, + "num_train_epochs": 2, + "save_steps": 50, + "total_flos": 1.6713924878742323e+17, + "trial_name": null, + "trial_params": null +} diff --git a/multi-prompt/twochar/checkpoint-200/training_args.bin b/multi-prompt/twochar/checkpoint-200/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..47d6200b8d4047ff364e6b15cf014dde61f2166b --- /dev/null +++ b/multi-prompt/twochar/checkpoint-200/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac3fb199292750362f1e387a5d8bde7088e81478bbe173194e512403378f1c70 +size 4475 diff --git a/multi-prompt/twochar/checkpoint-50/README.md b/multi-prompt/twochar/checkpoint-50/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/multi-prompt/twochar/checkpoint-50/adapter_config.json b/multi-prompt/twochar/checkpoint-50/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b33f8685004fc9f9eef2843be41eb6081508fc38 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.01, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "v_proj", + "q_proj", + "down_proj", + "k_proj", + "gate_proj", + "o_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/multi-prompt/twochar/checkpoint-50/adapter_model.bin b/multi-prompt/twochar/checkpoint-50/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..2f26f32c7a8c453fe666d94983d7c87659ca0240 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ff9f2ca982d4c94bf1984cebd603a043e51c0688b2c7ebbee14d46cf8dbcdcfe +size 125374989 diff --git a/multi-prompt/twochar/checkpoint-50/adapter_model/README.md b/multi-prompt/twochar/checkpoint-50/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/adapter_model/README.md @@ -0,0 +1,21 @@ +--- +library_name: peft +--- +## Training procedure + + +The following `bitsandbytes` quantization config was used during training: +- quant_method: bitsandbytes +- load_in_8bit: False +- load_in_4bit: True +- llm_int8_threshold: 6.0 +- llm_int8_skip_modules: None +- llm_int8_enable_fp32_cpu_offload: False +- llm_int8_has_fp16_weight: False +- bnb_4bit_quant_type: nf4 +- bnb_4bit_use_double_quant: True +- bnb_4bit_compute_dtype: bfloat16 +### Framework versions + + +- PEFT 0.6.0.dev0 diff --git a/multi-prompt/twochar/checkpoint-50/adapter_model/adapter_config.json b/multi-prompt/twochar/checkpoint-50/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..b33f8685004fc9f9eef2843be41eb6081508fc38 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/adapter_model/adapter_config.json @@ -0,0 +1,26 @@ +{ + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-2-13b-hf", + "bias": "none", + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layers_pattern": null, + "layers_to_transform": null, + "lora_alpha": 16, + "lora_dropout": 0.01, + "modules_to_save": null, + "peft_type": "LORA", + "r": 8, + "revision": null, + "target_modules": [ + "v_proj", + "q_proj", + "down_proj", + "k_proj", + "gate_proj", + "o_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/multi-prompt/twochar/checkpoint-50/adapter_model/adapter_model.bin b/multi-prompt/twochar/checkpoint-50/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..2f26f32c7a8c453fe666d94983d7c87659ca0240 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ff9f2ca982d4c94bf1984cebd603a043e51c0688b2c7ebbee14d46cf8dbcdcfe +size 125374989 diff --git a/multi-prompt/twochar/checkpoint-50/optimizer.pt b/multi-prompt/twochar/checkpoint-50/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..a8b935ad800ec9e7860de2a93f2c0ce470e20c71 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97bdb6008d49ae7a5e669f8a5688e4d26503776faaa8f016deecbb48fbd9e87c +size 250681597 diff --git a/multi-prompt/twochar/checkpoint-50/rng_state_0.pth b/multi-prompt/twochar/checkpoint-50/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..72b662466f58c7db850cfef57f82a2ec023746e4 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:378041dd0edab043964c17a1498bb1e053b53be09587a86c91983bfd003b402b +size 21687 diff --git a/multi-prompt/twochar/checkpoint-50/rng_state_1.pth b/multi-prompt/twochar/checkpoint-50/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..29f0804883c5a894698a7fc435db06826bfcd53c --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ae8625521971de986a3fe9a83426bc9be410a25486ff090d580bb201b91ac627 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-50/rng_state_2.pth b/multi-prompt/twochar/checkpoint-50/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..020aa958562074fb02dfd261fb7ab17ef540abf4 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a1b9c9c8dc3258484444dd2bf445247a35bc336f53a47204202a54fb6afbdc8d +size 21687 diff --git a/multi-prompt/twochar/checkpoint-50/rng_state_3.pth b/multi-prompt/twochar/checkpoint-50/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..a9b062ffdf8152b11bab4c9b856a417b48a99b10 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b94926ac56f32be2195666cf81c1c71845d6d44a1679442e46d2526336a6140 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-50/rng_state_4.pth b/multi-prompt/twochar/checkpoint-50/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..287217400ace09086b92ee95413967f4343b2498 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c78c64ffa806625bf637e6aa6b7774f099fb4e67041e78daaa625279d2694463 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-50/rng_state_5.pth b/multi-prompt/twochar/checkpoint-50/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..09c5fad4fd1b0e65f746fd42f0061479f2d56c11 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:76cb87e9b3de42221799a59f9f5e4a326327de746c8b9819ceaeea29b729fcb4 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-50/rng_state_6.pth b/multi-prompt/twochar/checkpoint-50/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..33b3442d82a5cce1b337a09cd59acf7ff173f268 --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:67f914005f2340a9fbb9ec4f599910b12669db318a4462202dd7dd60e50b2ca8 +size 21687 diff --git a/multi-prompt/twochar/checkpoint-50/rng_state_7.pth b/multi-prompt/twochar/checkpoint-50/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..ab863846afd96aa648917bcd70edac42a158c90e --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c7f956d6c91aadf25a203f3b37e60d03f812c2e0cf7591573e4c904f3da82f7d +size 21687 diff --git a/multi-prompt/twochar/checkpoint-50/scheduler.pt b/multi-prompt/twochar/checkpoint-50/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..052f2ada5cd173a143a172c13df4ada4c11c059a --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0e54688b73ae6a59a6bcf42b1a01810dc387a75e410ba415dabec8c9a74ae5f2 +size 627 diff --git a/multi-prompt/twochar/checkpoint-50/trainer_state.json b/multi-prompt/twochar/checkpoint-50/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..14281618ea239e82104de28bb3d6b1bcb49c1bdf --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/trainer_state.json @@ -0,0 +1,335 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.49504950495049505, + "eval_steps": 20, + "global_step": 50, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.01, + "learning_rate": 6.5e-06, + "loss": 2.0407, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 1.3e-05, + "loss": 2.3107, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 1.9499999999999996e-05, + "loss": 2.1613, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 2.6e-05, + "loss": 2.0444, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 3.25e-05, + "loss": 2.2512, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 3.899999999999999e-05, + "loss": 2.0516, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 4.5499999999999995e-05, + "loss": 2.2006, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 5.2e-05, + "loss": 2.1229, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 5.85e-05, + "loss": 2.2498, + "step": 9 + }, + { + "epoch": 0.1, + "learning_rate": 6.5e-05, + "loss": 2.0293, + "step": 10 + }, + { + "epoch": 0.11, + "learning_rate": 6.499564948206075e-05, + "loss": 2.1316, + "step": 11 + }, + { + "epoch": 0.12, + "learning_rate": 6.498259909298188e-05, + "loss": 2.1274, + "step": 12 + }, + { + "epoch": 0.13, + "learning_rate": 6.49608523266681e-05, + "loss": 2.1129, + "step": 13 + }, + { + "epoch": 0.14, + "learning_rate": 6.493041500525461e-05, + "loss": 2.1517, + "step": 14 + }, + { + "epoch": 0.15, + "learning_rate": 6.489129527754836e-05, + "loss": 2.1078, + "step": 15 + }, + { + "epoch": 0.16, + "learning_rate": 6.484350361684639e-05, + "loss": 2.0362, + "step": 16 + }, + { + "epoch": 0.17, + "learning_rate": 6.478705281813194e-05, + "loss": 2.0553, + "step": 17 + }, + { + "epoch": 0.18, + "learning_rate": 6.472195799464884e-05, + "loss": 2.1223, + "step": 18 + }, + { + "epoch": 0.19, + "learning_rate": 6.464823657385538e-05, + "loss": 2.0326, + "step": 19 + }, + { + "epoch": 0.2, + "learning_rate": 6.456590829275857e-05, + "loss": 2.1802, + "step": 20 + }, + { + "epoch": 0.2, + "eval_loss": 2.012301445007324, + "eval_runtime": 1.4855, + "eval_samples_per_second": 5.385, + "eval_steps_per_second": 0.673, + "step": 20 + }, + { + "epoch": 0.21, + "learning_rate": 6.447499519263001e-05, + "loss": 2.0665, + "step": 21 + }, + { + "epoch": 0.22, + "learning_rate": 6.437552161310498e-05, + "loss": 2.0354, + "step": 22 + }, + { + "epoch": 0.23, + "learning_rate": 6.426751418566609e-05, + "loss": 2.1289, + "step": 23 + }, + { + "epoch": 0.24, + "learning_rate": 6.415100182651334e-05, + "loss": 2.0742, + "step": 24 + }, + { + "epoch": 0.25, + "learning_rate": 6.402601572882268e-05, + "loss": 2.0412, + "step": 25 + }, + { + "epoch": 0.26, + "learning_rate": 6.38925893543947e-05, + "loss": 2.1974, + "step": 26 + }, + { + "epoch": 0.27, + "learning_rate": 6.375075842469626e-05, + "loss": 2.196, + "step": 27 + }, + { + "epoch": 0.28, + "learning_rate": 6.360056091129678e-05, + "loss": 2.0775, + "step": 28 + }, + { + "epoch": 0.29, + "learning_rate": 6.344203702570254e-05, + "loss": 2.0366, + "step": 29 + }, + { + "epoch": 0.3, + "learning_rate": 6.327522920859093e-05, + "loss": 2.1438, + "step": 30 + }, + { + "epoch": 0.31, + "learning_rate": 6.310018211844817e-05, + "loss": 2.0639, + "step": 31 + }, + { + "epoch": 0.32, + "learning_rate": 6.291694261961308e-05, + "loss": 2.1794, + "step": 32 + }, + { + "epoch": 0.33, + "learning_rate": 6.272555976973039e-05, + "loss": 2.0592, + "step": 33 + }, + { + "epoch": 0.34, + "learning_rate": 6.252608480661681e-05, + "loss": 2.1628, + "step": 34 + }, + { + "epoch": 0.35, + "learning_rate": 6.231857113454346e-05, + "loss": 2.1851, + "step": 35 + }, + { + "epoch": 0.36, + "learning_rate": 6.21030743099382e-05, + "loss": 2.0531, + "step": 36 + }, + { + "epoch": 0.37, + "learning_rate": 6.18796520265119e-05, + "loss": 2.0284, + "step": 37 + }, + { + "epoch": 0.38, + "learning_rate": 6.164836409981237e-05, + "loss": 2.2359, + "step": 38 + }, + { + "epoch": 0.39, + "learning_rate": 6.140927245121032e-05, + "loss": 1.8967, + "step": 39 + }, + { + "epoch": 0.4, + "learning_rate": 6.116244109132153e-05, + "loss": 2.0275, + "step": 40 + }, + { + "epoch": 0.4, + "eval_loss": 1.9861761331558228, + "eval_runtime": 1.4896, + "eval_samples_per_second": 5.37, + "eval_steps_per_second": 0.671, + "step": 40 + }, + { + "epoch": 0.41, + "learning_rate": 6.0907936102869656e-05, + "loss": 2.0316, + "step": 41 + }, + { + "epoch": 0.42, + "learning_rate": 6.0645825622994254e-05, + "loss": 2.0768, + "step": 42 + }, + { + "epoch": 0.43, + "learning_rate": 6.037617982500884e-05, + "loss": 2.0058, + "step": 43 + }, + { + "epoch": 0.44, + "learning_rate": 6.009907089961381e-05, + "loss": 2.2319, + "step": 44 + }, + { + "epoch": 0.45, + "learning_rate": 5.981457303556923e-05, + "loss": 2.1693, + "step": 45 + }, + { + "epoch": 0.46, + "learning_rate": 5.9522762399832716e-05, + "loss": 2.0858, + "step": 46 + }, + { + "epoch": 0.47, + "learning_rate": 5.922371711716768e-05, + "loss": 2.0576, + "step": 47 + }, + { + "epoch": 0.48, + "learning_rate": 5.891751724922749e-05, + "loss": 1.9946, + "step": 48 + }, + { + "epoch": 0.49, + "learning_rate": 5.860424477312095e-05, + "loss": 2.0166, + "step": 49 + }, + { + "epoch": 0.5, + "learning_rate": 5.828398355946514e-05, + "loss": 2.1482, + "step": 50 + } + ], + "logging_steps": 1, + "max_steps": 202, + "num_train_epochs": 2, + "save_steps": 50, + "total_flos": 4.187835255803085e+16, + "trial_name": null, + "trial_params": null +} diff --git a/multi-prompt/twochar/checkpoint-50/training_args.bin b/multi-prompt/twochar/checkpoint-50/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..47d6200b8d4047ff364e6b15cf014dde61f2166b --- /dev/null +++ b/multi-prompt/twochar/checkpoint-50/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac3fb199292750362f1e387a5d8bde7088e81478bbe173194e512403378f1c70 +size 4475 diff --git a/multi-prompt/twochar/config.json b/multi-prompt/twochar/config.json new file mode 100644 index 0000000000000000000000000000000000000000..99be7fbf147c96ef93ed720729ea90e8e21bdfca --- /dev/null +++ b/multi-prompt/twochar/config.json @@ -0,0 +1,38 @@ +{ + "_name_or_path": "meta-llama/Llama-2-13b-hf", + "architectures": [ + "LlamaForCausalLM" + ], + "bos_token_id": 1, + "eos_token_id": 2, + "hidden_act": "silu", + "hidden_size": 5120, + "initializer_range": 0.02, + "intermediate_size": 13824, + "max_position_embeddings": 4096, + "model_type": "llama", + "num_attention_heads": 40, + "num_hidden_layers": 40, + "num_key_value_heads": 40, + "pretraining_tp": 1, + "quantization_config": { + "bnb_4bit_compute_dtype": "bfloat16", + "bnb_4bit_quant_type": "nf4", + "bnb_4bit_use_double_quant": true, + "llm_int8_enable_fp32_cpu_offload": false, + "llm_int8_has_fp16_weight": false, + "llm_int8_skip_modules": null, + "llm_int8_threshold": 6.0, + "load_in_4bit": true, + "load_in_8bit": false, + "quant_method": "bitsandbytes" + }, + "rms_norm_eps": 1e-05, + "rope_scaling": null, + "rope_theta": 10000.0, + "tie_word_embeddings": false, + "torch_dtype": "float16", + "transformers_version": "4.34.0.dev0", + "use_cache": false, + "vocab_size": 32001 +} diff --git a/multi-prompt/twochar/special_tokens_map.json b/multi-prompt/twochar/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..3f58a5e115855c6ea3cec98accae196ad927222e --- /dev/null +++ b/multi-prompt/twochar/special_tokens_map.json @@ -0,0 +1,6 @@ +{ + "bos_token": "", + "eos_token": "", + "pad_token": "[PAD]", + "unk_token": "" +} diff --git a/multi-prompt/twochar/tokenizer.model b/multi-prompt/twochar/tokenizer.model new file mode 100644 index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899 --- /dev/null +++ b/multi-prompt/twochar/tokenizer.model @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347 +size 499723 diff --git a/multi-prompt/twochar/tokenizer_config.json b/multi-prompt/twochar/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..77849a9a3ca128f00bc3febdc9b39391a4ef43a3 --- /dev/null +++ b/multi-prompt/twochar/tokenizer_config.json @@ -0,0 +1,39 @@ +{ + "add_bos_token": true, + "add_eos_token": false, + "bos_token": { + "__type": "AddedToken", + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "clean_up_tokenization_spaces": false, + "eos_token": { + "__type": "AddedToken", + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "legacy": false, + "model_max_length": 1000000000000000019884624838656, + "pad_token": null, + "padding_side": "right", + "sp_model_kwargs": {}, + "spaces_between_special_tokens": false, + "tokenizer_class": "LlamaTokenizer", + "trust_remote_code": false, + "unk_token": { + "__type": "AddedToken", + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "use_default_system_prompt": true, + "use_fast": true +}