diff --git a/prompt-a/twochar/README.md b/prompt-a/twochar/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/prompt-a/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/prompt-a/twochar/adapter_config.json b/prompt-a/twochar/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..e15b2402bceaab5939e21de3eb1d9a6ad2980bbd --- /dev/null +++ b/prompt-a/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": [ + "down_proj", + "k_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/prompt-a/twochar/adapter_model.bin b/prompt-a/twochar/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..c96c513e5c63ebcef49c385fe35a3a5331b09141 --- /dev/null +++ b/prompt-a/twochar/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3686a6854a70ed7424cb6434eaf2d2d208980a99034b0d2075128de53b3297a2 +size 125374989 diff --git a/prompt-a/twochar/added_tokens.json b/prompt-a/twochar/added_tokens.json new file mode 100644 index 0000000000000000000000000000000000000000..e41416ddd79948246ea2dced6800ea3cd531c424 --- /dev/null +++ b/prompt-a/twochar/added_tokens.json @@ -0,0 +1,3 @@ +{ + "[PAD]": 32000 +} diff --git a/prompt-a/twochar/checkpoint-100/README.md b/prompt-a/twochar/checkpoint-100/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/prompt-a/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/prompt-a/twochar/checkpoint-100/adapter_config.json b/prompt-a/twochar/checkpoint-100/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..e15b2402bceaab5939e21de3eb1d9a6ad2980bbd --- /dev/null +++ b/prompt-a/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": [ + "down_proj", + "k_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/prompt-a/twochar/checkpoint-100/adapter_model.bin b/prompt-a/twochar/checkpoint-100/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..49151612708bf6f92b80633f78936fc9edd40e8f --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:42f483a8475f9c9615a60f6b89a84b8920c7a840df45ceabed0a3effadffc0e0 +size 125374989 diff --git a/prompt-a/twochar/checkpoint-100/adapter_model/README.md b/prompt-a/twochar/checkpoint-100/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/prompt-a/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/prompt-a/twochar/checkpoint-100/adapter_model/adapter_config.json b/prompt-a/twochar/checkpoint-100/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..e15b2402bceaab5939e21de3eb1d9a6ad2980bbd --- /dev/null +++ b/prompt-a/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": [ + "down_proj", + "k_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/prompt-a/twochar/checkpoint-100/adapter_model/adapter_model.bin b/prompt-a/twochar/checkpoint-100/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..49151612708bf6f92b80633f78936fc9edd40e8f --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:42f483a8475f9c9615a60f6b89a84b8920c7a840df45ceabed0a3effadffc0e0 +size 125374989 diff --git a/prompt-a/twochar/checkpoint-100/optimizer.pt b/prompt-a/twochar/checkpoint-100/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..1c5d9592c072640fc7938faa7a0d46ccca79bd4f --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:20d2e41d040cc30ba3448d2ef29ddff84499af170be7c5a74b063a9a719180ad +size 250681597 diff --git a/prompt-a/twochar/checkpoint-100/rng_state_0.pth b/prompt-a/twochar/checkpoint-100/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..ad9269ebf081152ae5eb5ab3ee652a4345bb7e78 --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2f4ac5c72b3a8d37d6f5f4e72f59a981105e0739be3b7567447236a07702ef98 +size 21687 diff --git a/prompt-a/twochar/checkpoint-100/rng_state_1.pth b/prompt-a/twochar/checkpoint-100/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..2c25cd2c6610ab0b0ee6505778c9f72c0e778929 --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:09a9e7711a12aff1cb01365952914663e874439fc80ca6174d7f1ed92dd1fd03 +size 21687 diff --git a/prompt-a/twochar/checkpoint-100/rng_state_2.pth b/prompt-a/twochar/checkpoint-100/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..37cc9e20eb59043d72cead19ce5dc45258cc2636 --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e71e7c3877097d7a81872b94110dacc2bb9179e8811b27061b892815e8d9aa49 +size 21687 diff --git a/prompt-a/twochar/checkpoint-100/rng_state_3.pth b/prompt-a/twochar/checkpoint-100/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..6192a3dd961ecf758de8e77528077bd1e7222f9e --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:101962960d17908eccf4d72170394b7c436c70f3676372f4a98047973af8b216 +size 21687 diff --git a/prompt-a/twochar/checkpoint-100/rng_state_4.pth b/prompt-a/twochar/checkpoint-100/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..78a4c2caca665f09924545c8d9ec7b9ee70af367 --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3419ae90d4ffed73c9b8458beb5e60e1e5a986b8daf2d55fd3289f908f2ef5c9 +size 21687 diff --git a/prompt-a/twochar/checkpoint-100/rng_state_5.pth b/prompt-a/twochar/checkpoint-100/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..8f7687a1d24f89e63e2a757a192c6bf93d76ba97 --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:064ab007d23c2b923f5dce2324188582329cc05945abb2006e33f8fbb94f27e0 +size 21687 diff --git a/prompt-a/twochar/checkpoint-100/rng_state_6.pth b/prompt-a/twochar/checkpoint-100/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..063ab9b97b6e3c347ec39d2f85b182f0ccc5f3c6 --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a231d16fbc0b8c521a5b6329b1042c9237bbaaf2cfddf194cc2c875474f5d4a9 +size 21687 diff --git a/prompt-a/twochar/checkpoint-100/rng_state_7.pth b/prompt-a/twochar/checkpoint-100/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..26015a28868cef53a04382443b335393ceb0c07f --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ab804b08c8130b997416c537045d5e5f947a41304867b421f39443b0b80131fe +size 21687 diff --git a/prompt-a/twochar/checkpoint-100/scheduler.pt b/prompt-a/twochar/checkpoint-100/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..4d832c94e2de61db5216a37122c6fd1112328631 --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b066d96ed0b5d7ac788c3a38bca16c4cd9e92b1f62ff59691c8a4867593fd5cf +size 627 diff --git a/prompt-a/twochar/checkpoint-100/trainer_state.json b/prompt-a/twochar/checkpoint-100/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..6df5ae3c271883edf90d3a885b5442ba1369420b --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/trainer_state.json @@ -0,0 +1,659 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.0, + "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": 1.9562, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 1.3e-05, + "loss": 2.0273, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 1.9499999999999996e-05, + "loss": 2.2516, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 2.6e-05, + "loss": 2.1955, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 3.25e-05, + "loss": 2.1752, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 3.899999999999999e-05, + "loss": 2.2144, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 4.5499999999999995e-05, + "loss": 2.0829, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 5.2e-05, + "loss": 2.0503, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 5.85e-05, + "loss": 2.1558, + "step": 9 + }, + { + "epoch": 0.1, + "learning_rate": 6.5e-05, + "loss": 2.0924, + "step": 10 + }, + { + "epoch": 0.11, + "learning_rate": 6.499555741225524e-05, + "loss": 2.1238, + "step": 11 + }, + { + "epoch": 0.12, + "learning_rate": 6.498223086358009e-05, + "loss": 2.129, + "step": 12 + }, + { + "epoch": 0.13, + "learning_rate": 6.49600239973199e-05, + "loss": 2.0559, + "step": 13 + }, + { + "epoch": 0.14, + "learning_rate": 6.492894288461016e-05, + "loss": 2.0659, + "step": 14 + }, + { + "epoch": 0.15, + "learning_rate": 6.488899602271677e-05, + "loss": 2.1563, + "step": 15 + }, + { + "epoch": 0.16, + "learning_rate": 6.484019433271287e-05, + "loss": 2.1706, + "step": 16 + }, + { + "epoch": 0.17, + "learning_rate": 6.478255115649324e-05, + "loss": 2.0736, + "step": 17 + }, + { + "epoch": 0.18, + "learning_rate": 6.471608225312668e-05, + "loss": 2.1295, + "step": 18 + }, + { + "epoch": 0.19, + "learning_rate": 6.464080579454772e-05, + "loss": 2.0618, + "step": 19 + }, + { + "epoch": 0.2, + "learning_rate": 6.455674236058847e-05, + "loss": 2.12, + "step": 20 + }, + { + "epoch": 0.2, + "eval_loss": 2.1955978870391846, + "eval_runtime": 2.9627, + "eval_samples_per_second": 3.375, + "eval_steps_per_second": 0.675, + "step": 20 + }, + { + "epoch": 0.21, + "learning_rate": 6.446391493335242e-05, + "loss": 2.1576, + "step": 21 + }, + { + "epoch": 0.22, + "learning_rate": 6.43623488909313e-05, + "loss": 2.0446, + "step": 22 + }, + { + "epoch": 0.23, + "learning_rate": 6.425207200046697e-05, + "loss": 2.0033, + "step": 23 + }, + { + "epoch": 0.24, + "learning_rate": 6.413311441056019e-05, + "loss": 2.1382, + "step": 24 + }, + { + "epoch": 0.25, + "learning_rate": 6.400550864302823e-05, + "loss": 2.0989, + "step": 25 + }, + { + "epoch": 0.26, + "learning_rate": 6.386928958401383e-05, + "loss": 2.1276, + "step": 26 + }, + { + "epoch": 0.27, + "learning_rate": 6.372449447444757e-05, + "loss": 2.023, + "step": 27 + }, + { + "epoch": 0.28, + "learning_rate": 6.357116289986662e-05, + "loss": 2.0597, + "step": 28 + }, + { + "epoch": 0.29, + "learning_rate": 6.340933677959248e-05, + "loss": 2.0529, + "step": 29 + }, + { + "epoch": 0.3, + "learning_rate": 6.323906035527062e-05, + "loss": 1.9873, + "step": 30 + }, + { + "epoch": 0.31, + "learning_rate": 6.306038017877523e-05, + "loss": 2.0613, + "step": 31 + }, + { + "epoch": 0.32, + "learning_rate": 6.287334509948247e-05, + "loss": 2.0765, + "step": 32 + }, + { + "epoch": 0.33, + "learning_rate": 6.267800625091547e-05, + "loss": 2.0809, + "step": 33 + }, + { + "epoch": 0.34, + "learning_rate": 6.2474417036765e-05, + "loss": 2.072, + "step": 34 + }, + { + "epoch": 0.35, + "learning_rate": 6.226263311628936e-05, + "loss": 2.0069, + "step": 35 + }, + { + "epoch": 0.36, + "learning_rate": 6.204271238909777e-05, + "loss": 2.0784, + "step": 36 + }, + { + "epoch": 0.37, + "learning_rate": 6.181471497932117e-05, + "loss": 2.1763, + "step": 37 + }, + { + "epoch": 0.38, + "learning_rate": 6.157870321917481e-05, + "loss": 2.0402, + "step": 38 + }, + { + "epoch": 0.39, + "learning_rate": 6.133474163191738e-05, + "loss": 2.1082, + "step": 39 + }, + { + "epoch": 0.4, + "learning_rate": 6.108289691421088e-05, + "loss": 2.1381, + "step": 40 + }, + { + "epoch": 0.4, + "eval_loss": 2.172107219696045, + "eval_runtime": 2.97, + "eval_samples_per_second": 3.367, + "eval_steps_per_second": 0.673, + "step": 40 + }, + { + "epoch": 0.41, + "learning_rate": 6.08232379178865e-05, + "loss": 2.0538, + "step": 41 + }, + { + "epoch": 0.42, + "learning_rate": 6.055583563112115e-05, + "loss": 2.222, + "step": 42 + }, + { + "epoch": 0.43, + "learning_rate": 6.0280763159030015e-05, + "loss": 1.996, + "step": 43 + }, + { + "epoch": 0.44, + "learning_rate": 5.999809570368039e-05, + "loss": 1.9798, + "step": 44 + }, + { + "epoch": 0.45, + "learning_rate": 5.970791054353217e-05, + "loss": 2.1825, + "step": 45 + }, + { + "epoch": 0.46, + "learning_rate": 5.941028701231065e-05, + "loss": 2.1671, + "step": 46 + }, + { + "epoch": 0.47, + "learning_rate": 5.91053064773175e-05, + "loss": 2.0992, + "step": 47 + }, + { + "epoch": 0.48, + "learning_rate": 5.8793052317185785e-05, + "loss": 2.0544, + "step": 48 + }, + { + "epoch": 0.49, + "learning_rate": 5.8473609899085046e-05, + "loss": 2.1054, + "step": 49 + }, + { + "epoch": 0.5, + "learning_rate": 5.814706655538279e-05, + "loss": 2.1214, + "step": 50 + }, + { + "epoch": 0.51, + "learning_rate": 5.781351155976865e-05, + "loss": 1.9961, + "step": 51 + }, + { + "epoch": 0.52, + "learning_rate": 5.7473036102847915e-05, + "loss": 2.0233, + "step": 52 + }, + { + "epoch": 0.53, + "learning_rate": 5.712573326721088e-05, + "loss": 2.0756, + "step": 53 + }, + { + "epoch": 0.54, + "learning_rate": 5.6771698001984996e-05, + "loss": 2.0138, + "step": 54 + }, + { + "epoch": 0.55, + "learning_rate": 5.641102709687678e-05, + "loss": 1.9964, + "step": 55 + }, + { + "epoch": 0.56, + "learning_rate": 5.604381915571039e-05, + "loss": 2.0915, + "step": 56 + }, + { + "epoch": 0.57, + "learning_rate": 5.5670174569470405e-05, + "loss": 1.9954, + "step": 57 + }, + { + "epoch": 0.58, + "learning_rate": 5.529019548885587e-05, + "loss": 2.1395, + "step": 58 + }, + { + "epoch": 0.59, + "learning_rate": 5.490398579635338e-05, + "loss": 2.0388, + "step": 59 + }, + { + "epoch": 0.6, + "learning_rate": 5.451165107783658e-05, + "loss": 2.0979, + "step": 60 + }, + { + "epoch": 0.6, + "eval_loss": 2.166550874710083, + "eval_runtime": 2.9777, + "eval_samples_per_second": 3.358, + "eval_steps_per_second": 0.672, + "step": 60 + }, + { + "epoch": 0.61, + "learning_rate": 5.411329859370006e-05, + "loss": 2.0614, + "step": 61 + }, + { + "epoch": 0.62, + "learning_rate": 5.370903724953548e-05, + "loss": 2.061, + "step": 62 + }, + { + "epoch": 0.63, + "learning_rate": 5.329897756635779e-05, + "loss": 2.0868, + "step": 63 + }, + { + "epoch": 0.64, + "learning_rate": 5.2883231650389946e-05, + "loss": 2.0659, + "step": 64 + }, + { + "epoch": 0.65, + "learning_rate": 5.2461913162414206e-05, + "loss": 2.1325, + "step": 65 + }, + { + "epoch": 0.66, + "learning_rate": 5.203513728669831e-05, + "loss": 2.0955, + "step": 66 + }, + { + "epoch": 0.67, + "learning_rate": 5.160302069950537e-05, + "loss": 1.9692, + "step": 67 + }, + { + "epoch": 0.68, + "learning_rate": 5.116568153719566e-05, + "loss": 2.0858, + "step": 68 + }, + { + "epoch": 0.69, + "learning_rate": 5.072323936392936e-05, + "loss": 2.0884, + "step": 69 + }, + { + "epoch": 0.7, + "learning_rate": 5.027581513897888e-05, + "loss": 1.952, + "step": 70 + }, + { + "epoch": 0.71, + "learning_rate": 4.9823531183659796e-05, + "loss": 2.0223, + "step": 71 + }, + { + "epoch": 0.72, + "learning_rate": 4.936651114788949e-05, + "loss": 2.0098, + "step": 72 + }, + { + "epoch": 0.73, + "learning_rate": 4.890487997638242e-05, + "loss": 1.9965, + "step": 73 + }, + { + "epoch": 0.74, + "learning_rate": 4.843876387449151e-05, + "loss": 2.1097, + "step": 74 + }, + { + "epoch": 0.75, + "learning_rate": 4.796829027370489e-05, + "loss": 2.1051, + "step": 75 + }, + { + "epoch": 0.76, + "learning_rate": 4.749358779680735e-05, + "loss": 2.1672, + "step": 76 + }, + { + "epoch": 0.77, + "learning_rate": 4.701478622271621e-05, + "loss": 2.0641, + "step": 77 + }, + { + "epoch": 0.78, + "learning_rate": 4.653201645100101e-05, + "loss": 2.1899, + "step": 78 + }, + { + "epoch": 0.79, + "learning_rate": 4.6045410466096934e-05, + "loss": 2.007, + "step": 79 + }, + { + "epoch": 0.8, + "learning_rate": 4.555510130122151e-05, + "loss": 2.1355, + "step": 80 + }, + { + "epoch": 0.8, + "eval_loss": 2.160327434539795, + "eval_runtime": 2.9723, + "eval_samples_per_second": 3.364, + "eval_steps_per_second": 0.673, + "step": 80 + }, + { + "epoch": 0.81, + "learning_rate": 4.5061223002004694e-05, + "loss": 2.0046, + "step": 81 + }, + { + "epoch": 0.82, + "learning_rate": 4.456391058984217e-05, + "loss": 2.0902, + "step": 82 + }, + { + "epoch": 0.83, + "learning_rate": 4.406330002498178e-05, + "loss": 2.1259, + "step": 83 + }, + { + "epoch": 0.84, + "learning_rate": 4.3559528169353394e-05, + "loss": 1.9241, + "step": 84 + }, + { + "epoch": 0.85, + "learning_rate": 4.3052732749152215e-05, + "loss": 2.1537, + "step": 85 + }, + { + "epoch": 0.86, + "learning_rate": 4.254305231718579e-05, + "loss": 1.9482, + "step": 86 + }, + { + "epoch": 0.87, + "learning_rate": 4.2030626214995104e-05, + "loss": 2.0181, + "step": 87 + }, + { + "epoch": 0.88, + "learning_rate": 4.151559453475994e-05, + "loss": 1.9839, + "step": 88 + }, + { + "epoch": 0.89, + "learning_rate": 4.0998098080999136e-05, + "loss": 2.0735, + "step": 89 + }, + { + "epoch": 0.9, + "learning_rate": 4.047827833207597e-05, + "loss": 2.1691, + "step": 90 + }, + { + "epoch": 0.91, + "learning_rate": 3.995627740151942e-05, + "loss": 2.2761, + "step": 91 + }, + { + "epoch": 0.92, + "learning_rate": 3.9432237999171774e-05, + "loss": 2.1657, + "step": 92 + }, + { + "epoch": 0.93, + "learning_rate": 3.89063033921731e-05, + "loss": 1.9501, + "step": 93 + }, + { + "epoch": 0.94, + "learning_rate": 3.8378617365793576e-05, + "loss": 2.0869, + "step": 94 + }, + { + "epoch": 0.95, + "learning_rate": 3.784932418412385e-05, + "loss": 2.0112, + "step": 95 + }, + { + "epoch": 0.96, + "learning_rate": 3.731856855063486e-05, + "loss": 1.9212, + "step": 96 + }, + { + "epoch": 0.97, + "learning_rate": 3.678649556861724e-05, + "loss": 2.0395, + "step": 97 + }, + { + "epoch": 0.98, + "learning_rate": 3.625325070151155e-05, + "loss": 2.0179, + "step": 98 + }, + { + "epoch": 0.99, + "learning_rate": 3.571897973314004e-05, + "loss": 1.9713, + "step": 99 + }, + { + "epoch": 1.0, + "learning_rate": 3.51838287278508e-05, + "loss": 2.0756, + "step": 100 + }, + { + "epoch": 1.0, + "eval_loss": 2.1577982902526855, + "eval_runtime": 2.9721, + "eval_samples_per_second": 3.365, + "eval_steps_per_second": 0.673, + "step": 100 + } + ], + "logging_steps": 1, + "max_steps": 200, + "num_train_epochs": 2, + "save_steps": 50, + "total_flos": 8.361608023638016e+16, + "trial_name": null, + "trial_params": null +} diff --git a/prompt-a/twochar/checkpoint-100/training_args.bin b/prompt-a/twochar/checkpoint-100/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..d58072f53f91b75b135f9498c19f7bcc5b637ba4 --- /dev/null +++ b/prompt-a/twochar/checkpoint-100/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9212c1ed1ebf30dd523c0f488bee8e1daf66e67eaaf76d60029260d57f02f88 +size 4411 diff --git a/prompt-a/twochar/checkpoint-150/README.md b/prompt-a/twochar/checkpoint-150/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/prompt-a/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/prompt-a/twochar/checkpoint-150/adapter_config.json b/prompt-a/twochar/checkpoint-150/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..e15b2402bceaab5939e21de3eb1d9a6ad2980bbd --- /dev/null +++ b/prompt-a/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": [ + "down_proj", + "k_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/prompt-a/twochar/checkpoint-150/adapter_model.bin b/prompt-a/twochar/checkpoint-150/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..8a8efd52487c5a6d03c2acbd87646121a3bc8993 --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef2308923705e0db072fa56c1601591d86bd453af7564a28a77b84dd1d79e8c4 +size 125374989 diff --git a/prompt-a/twochar/checkpoint-150/adapter_model/README.md b/prompt-a/twochar/checkpoint-150/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/prompt-a/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/prompt-a/twochar/checkpoint-150/adapter_model/adapter_config.json b/prompt-a/twochar/checkpoint-150/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..e15b2402bceaab5939e21de3eb1d9a6ad2980bbd --- /dev/null +++ b/prompt-a/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": [ + "down_proj", + "k_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/prompt-a/twochar/checkpoint-150/adapter_model/adapter_model.bin b/prompt-a/twochar/checkpoint-150/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..8a8efd52487c5a6d03c2acbd87646121a3bc8993 --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef2308923705e0db072fa56c1601591d86bd453af7564a28a77b84dd1d79e8c4 +size 125374989 diff --git a/prompt-a/twochar/checkpoint-150/optimizer.pt b/prompt-a/twochar/checkpoint-150/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..bee1c631d6e08e9d85a2b0ea3676903b5231b59f --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:602ac917af5d4639ca282f520dde56e364bf3f0db429735790992bf32ae66b1d +size 250681597 diff --git a/prompt-a/twochar/checkpoint-150/rng_state_0.pth b/prompt-a/twochar/checkpoint-150/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..ff97c3eec33e99bff98915ef9653fc357a69a596 --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:89eadc3c165657fd8751b85f9ff44d7c95cc36af0669fbff4a4ffe5e0b07f5fc +size 21687 diff --git a/prompt-a/twochar/checkpoint-150/rng_state_1.pth b/prompt-a/twochar/checkpoint-150/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..f3906a31dca8232805cf57d482767b629b9edddc --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3745b667b2a6ca4a67753c6c91a46d6482d85c16a72b363be05fc59f7c442c14 +size 21687 diff --git a/prompt-a/twochar/checkpoint-150/rng_state_2.pth b/prompt-a/twochar/checkpoint-150/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..664d72b29c93bfd26f452d7040f3a706a2e36bfc --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:730ab402a78e7c0ffa824fcd5f320e6ac0fca6070a3982c84cac9eb233003d70 +size 21687 diff --git a/prompt-a/twochar/checkpoint-150/rng_state_3.pth b/prompt-a/twochar/checkpoint-150/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..6edda0af9099e51fc25e909f0025f262c63fc11d --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d6ee13cc98624b772f8f4143d6dccdc5cdaff35faa989b04753b6d57e692fa4c +size 21687 diff --git a/prompt-a/twochar/checkpoint-150/rng_state_4.pth b/prompt-a/twochar/checkpoint-150/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..4bd710f04455a879d85408cba5732797e8ab834b --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c7a46010650ea94fbb45a5cb40a5d7a9cc9a325ad0dcbbe32f26384cfedc3c9 +size 21687 diff --git a/prompt-a/twochar/checkpoint-150/rng_state_5.pth b/prompt-a/twochar/checkpoint-150/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..cf7573cbe764b170b82cd4319cf5c6b97f96f4c9 --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:448b2bc0d10409f11bb41fd0844a35fce476a3ffaa24967c8dfa611c6f55547d +size 21687 diff --git a/prompt-a/twochar/checkpoint-150/rng_state_6.pth b/prompt-a/twochar/checkpoint-150/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..135530c5623bf8d5ed5a1849641cde2073f0b2b1 --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a2d6cd239229ac1dd36dd86fc7bed222b898f45d69914c4668c12378f553f4be +size 21687 diff --git a/prompt-a/twochar/checkpoint-150/rng_state_7.pth b/prompt-a/twochar/checkpoint-150/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..9fd61e90c186fc30d0299e9d2d5b6850cc3cc878 --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d1f9bd38de615aa1185c1abc88320d66fe7f03cb4281a215b715cb5b31205b2e +size 21687 diff --git a/prompt-a/twochar/checkpoint-150/scheduler.pt b/prompt-a/twochar/checkpoint-150/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..f96a2443802cb8f1ec8dd6c856c6222f3dc2e216 --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:31da033238aea895275b0ce325aba2b20a03c1ced0b3d0cc962cb556ea08a8a9 +size 627 diff --git a/prompt-a/twochar/checkpoint-150/trainer_state.json b/prompt-a/twochar/checkpoint-150/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..546bb8ad60dcf9792311556f8d8d895fd1021f81 --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/trainer_state.json @@ -0,0 +1,975 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.5, + "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": 1.9562, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 1.3e-05, + "loss": 2.0273, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 1.9499999999999996e-05, + "loss": 2.2516, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 2.6e-05, + "loss": 2.1955, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 3.25e-05, + "loss": 2.1752, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 3.899999999999999e-05, + "loss": 2.2144, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 4.5499999999999995e-05, + "loss": 2.0829, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 5.2e-05, + "loss": 2.0503, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 5.85e-05, + "loss": 2.1558, + "step": 9 + }, + { + "epoch": 0.1, + "learning_rate": 6.5e-05, + "loss": 2.0924, + "step": 10 + }, + { + "epoch": 0.11, + "learning_rate": 6.499555741225524e-05, + "loss": 2.1238, + "step": 11 + }, + { + "epoch": 0.12, + "learning_rate": 6.498223086358009e-05, + "loss": 2.129, + "step": 12 + }, + { + "epoch": 0.13, + "learning_rate": 6.49600239973199e-05, + "loss": 2.0559, + "step": 13 + }, + { + "epoch": 0.14, + "learning_rate": 6.492894288461016e-05, + "loss": 2.0659, + "step": 14 + }, + { + "epoch": 0.15, + "learning_rate": 6.488899602271677e-05, + "loss": 2.1563, + "step": 15 + }, + { + "epoch": 0.16, + "learning_rate": 6.484019433271287e-05, + "loss": 2.1706, + "step": 16 + }, + { + "epoch": 0.17, + "learning_rate": 6.478255115649324e-05, + "loss": 2.0736, + "step": 17 + }, + { + "epoch": 0.18, + "learning_rate": 6.471608225312668e-05, + "loss": 2.1295, + "step": 18 + }, + { + "epoch": 0.19, + "learning_rate": 6.464080579454772e-05, + "loss": 2.0618, + "step": 19 + }, + { + "epoch": 0.2, + "learning_rate": 6.455674236058847e-05, + "loss": 2.12, + "step": 20 + }, + { + "epoch": 0.2, + "eval_loss": 2.1955978870391846, + "eval_runtime": 2.9627, + "eval_samples_per_second": 3.375, + "eval_steps_per_second": 0.675, + "step": 20 + }, + { + "epoch": 0.21, + "learning_rate": 6.446391493335242e-05, + "loss": 2.1576, + "step": 21 + }, + { + "epoch": 0.22, + "learning_rate": 6.43623488909313e-05, + "loss": 2.0446, + "step": 22 + }, + { + "epoch": 0.23, + "learning_rate": 6.425207200046697e-05, + "loss": 2.0033, + "step": 23 + }, + { + "epoch": 0.24, + "learning_rate": 6.413311441056019e-05, + "loss": 2.1382, + "step": 24 + }, + { + "epoch": 0.25, + "learning_rate": 6.400550864302823e-05, + "loss": 2.0989, + "step": 25 + }, + { + "epoch": 0.26, + "learning_rate": 6.386928958401383e-05, + "loss": 2.1276, + "step": 26 + }, + { + "epoch": 0.27, + "learning_rate": 6.372449447444757e-05, + "loss": 2.023, + "step": 27 + }, + { + "epoch": 0.28, + "learning_rate": 6.357116289986662e-05, + "loss": 2.0597, + "step": 28 + }, + { + "epoch": 0.29, + "learning_rate": 6.340933677959248e-05, + "loss": 2.0529, + "step": 29 + }, + { + "epoch": 0.3, + "learning_rate": 6.323906035527062e-05, + "loss": 1.9873, + "step": 30 + }, + { + "epoch": 0.31, + "learning_rate": 6.306038017877523e-05, + "loss": 2.0613, + "step": 31 + }, + { + "epoch": 0.32, + "learning_rate": 6.287334509948247e-05, + "loss": 2.0765, + "step": 32 + }, + { + "epoch": 0.33, + "learning_rate": 6.267800625091547e-05, + "loss": 2.0809, + "step": 33 + }, + { + "epoch": 0.34, + "learning_rate": 6.2474417036765e-05, + "loss": 2.072, + "step": 34 + }, + { + "epoch": 0.35, + "learning_rate": 6.226263311628936e-05, + "loss": 2.0069, + "step": 35 + }, + { + "epoch": 0.36, + "learning_rate": 6.204271238909777e-05, + "loss": 2.0784, + "step": 36 + }, + { + "epoch": 0.37, + "learning_rate": 6.181471497932117e-05, + "loss": 2.1763, + "step": 37 + }, + { + "epoch": 0.38, + "learning_rate": 6.157870321917481e-05, + "loss": 2.0402, + "step": 38 + }, + { + "epoch": 0.39, + "learning_rate": 6.133474163191738e-05, + "loss": 2.1082, + "step": 39 + }, + { + "epoch": 0.4, + "learning_rate": 6.108289691421088e-05, + "loss": 2.1381, + "step": 40 + }, + { + "epoch": 0.4, + "eval_loss": 2.172107219696045, + "eval_runtime": 2.97, + "eval_samples_per_second": 3.367, + "eval_steps_per_second": 0.673, + "step": 40 + }, + { + "epoch": 0.41, + "learning_rate": 6.08232379178865e-05, + "loss": 2.0538, + "step": 41 + }, + { + "epoch": 0.42, + "learning_rate": 6.055583563112115e-05, + "loss": 2.222, + "step": 42 + }, + { + "epoch": 0.43, + "learning_rate": 6.0280763159030015e-05, + "loss": 1.996, + "step": 43 + }, + { + "epoch": 0.44, + "learning_rate": 5.999809570368039e-05, + "loss": 1.9798, + "step": 44 + }, + { + "epoch": 0.45, + "learning_rate": 5.970791054353217e-05, + "loss": 2.1825, + "step": 45 + }, + { + "epoch": 0.46, + "learning_rate": 5.941028701231065e-05, + "loss": 2.1671, + "step": 46 + }, + { + "epoch": 0.47, + "learning_rate": 5.91053064773175e-05, + "loss": 2.0992, + "step": 47 + }, + { + "epoch": 0.48, + "learning_rate": 5.8793052317185785e-05, + "loss": 2.0544, + "step": 48 + }, + { + "epoch": 0.49, + "learning_rate": 5.8473609899085046e-05, + "loss": 2.1054, + "step": 49 + }, + { + "epoch": 0.5, + "learning_rate": 5.814706655538279e-05, + "loss": 2.1214, + "step": 50 + }, + { + "epoch": 0.51, + "learning_rate": 5.781351155976865e-05, + "loss": 1.9961, + "step": 51 + }, + { + "epoch": 0.52, + "learning_rate": 5.7473036102847915e-05, + "loss": 2.0233, + "step": 52 + }, + { + "epoch": 0.53, + "learning_rate": 5.712573326721088e-05, + "loss": 2.0756, + "step": 53 + }, + { + "epoch": 0.54, + "learning_rate": 5.6771698001984996e-05, + "loss": 2.0138, + "step": 54 + }, + { + "epoch": 0.55, + "learning_rate": 5.641102709687678e-05, + "loss": 1.9964, + "step": 55 + }, + { + "epoch": 0.56, + "learning_rate": 5.604381915571039e-05, + "loss": 2.0915, + "step": 56 + }, + { + "epoch": 0.57, + "learning_rate": 5.5670174569470405e-05, + "loss": 1.9954, + "step": 57 + }, + { + "epoch": 0.58, + "learning_rate": 5.529019548885587e-05, + "loss": 2.1395, + "step": 58 + }, + { + "epoch": 0.59, + "learning_rate": 5.490398579635338e-05, + "loss": 2.0388, + "step": 59 + }, + { + "epoch": 0.6, + "learning_rate": 5.451165107783658e-05, + "loss": 2.0979, + "step": 60 + }, + { + "epoch": 0.6, + "eval_loss": 2.166550874710083, + "eval_runtime": 2.9777, + "eval_samples_per_second": 3.358, + "eval_steps_per_second": 0.672, + "step": 60 + }, + { + "epoch": 0.61, + "learning_rate": 5.411329859370006e-05, + "loss": 2.0614, + "step": 61 + }, + { + "epoch": 0.62, + "learning_rate": 5.370903724953548e-05, + "loss": 2.061, + "step": 62 + }, + { + "epoch": 0.63, + "learning_rate": 5.329897756635779e-05, + "loss": 2.0868, + "step": 63 + }, + { + "epoch": 0.64, + "learning_rate": 5.2883231650389946e-05, + "loss": 2.0659, + "step": 64 + }, + { + "epoch": 0.65, + "learning_rate": 5.2461913162414206e-05, + "loss": 2.1325, + "step": 65 + }, + { + "epoch": 0.66, + "learning_rate": 5.203513728669831e-05, + "loss": 2.0955, + "step": 66 + }, + { + "epoch": 0.67, + "learning_rate": 5.160302069950537e-05, + "loss": 1.9692, + "step": 67 + }, + { + "epoch": 0.68, + "learning_rate": 5.116568153719566e-05, + "loss": 2.0858, + "step": 68 + }, + { + "epoch": 0.69, + "learning_rate": 5.072323936392936e-05, + "loss": 2.0884, + "step": 69 + }, + { + "epoch": 0.7, + "learning_rate": 5.027581513897888e-05, + "loss": 1.952, + "step": 70 + }, + { + "epoch": 0.71, + "learning_rate": 4.9823531183659796e-05, + "loss": 2.0223, + "step": 71 + }, + { + "epoch": 0.72, + "learning_rate": 4.936651114788949e-05, + "loss": 2.0098, + "step": 72 + }, + { + "epoch": 0.73, + "learning_rate": 4.890487997638242e-05, + "loss": 1.9965, + "step": 73 + }, + { + "epoch": 0.74, + "learning_rate": 4.843876387449151e-05, + "loss": 2.1097, + "step": 74 + }, + { + "epoch": 0.75, + "learning_rate": 4.796829027370489e-05, + "loss": 2.1051, + "step": 75 + }, + { + "epoch": 0.76, + "learning_rate": 4.749358779680735e-05, + "loss": 2.1672, + "step": 76 + }, + { + "epoch": 0.77, + "learning_rate": 4.701478622271621e-05, + "loss": 2.0641, + "step": 77 + }, + { + "epoch": 0.78, + "learning_rate": 4.653201645100101e-05, + "loss": 2.1899, + "step": 78 + }, + { + "epoch": 0.79, + "learning_rate": 4.6045410466096934e-05, + "loss": 2.007, + "step": 79 + }, + { + "epoch": 0.8, + "learning_rate": 4.555510130122151e-05, + "loss": 2.1355, + "step": 80 + }, + { + "epoch": 0.8, + "eval_loss": 2.160327434539795, + "eval_runtime": 2.9723, + "eval_samples_per_second": 3.364, + "eval_steps_per_second": 0.673, + "step": 80 + }, + { + "epoch": 0.81, + "learning_rate": 4.5061223002004694e-05, + "loss": 2.0046, + "step": 81 + }, + { + "epoch": 0.82, + "learning_rate": 4.456391058984217e-05, + "loss": 2.0902, + "step": 82 + }, + { + "epoch": 0.83, + "learning_rate": 4.406330002498178e-05, + "loss": 2.1259, + "step": 83 + }, + { + "epoch": 0.84, + "learning_rate": 4.3559528169353394e-05, + "loss": 1.9241, + "step": 84 + }, + { + "epoch": 0.85, + "learning_rate": 4.3052732749152215e-05, + "loss": 2.1537, + "step": 85 + }, + { + "epoch": 0.86, + "learning_rate": 4.254305231718579e-05, + "loss": 1.9482, + "step": 86 + }, + { + "epoch": 0.87, + "learning_rate": 4.2030626214995104e-05, + "loss": 2.0181, + "step": 87 + }, + { + "epoch": 0.88, + "learning_rate": 4.151559453475994e-05, + "loss": 1.9839, + "step": 88 + }, + { + "epoch": 0.89, + "learning_rate": 4.0998098080999136e-05, + "loss": 2.0735, + "step": 89 + }, + { + "epoch": 0.9, + "learning_rate": 4.047827833207597e-05, + "loss": 2.1691, + "step": 90 + }, + { + "epoch": 0.91, + "learning_rate": 3.995627740151942e-05, + "loss": 2.2761, + "step": 91 + }, + { + "epoch": 0.92, + "learning_rate": 3.9432237999171774e-05, + "loss": 2.1657, + "step": 92 + }, + { + "epoch": 0.93, + "learning_rate": 3.89063033921731e-05, + "loss": 1.9501, + "step": 93 + }, + { + "epoch": 0.94, + "learning_rate": 3.8378617365793576e-05, + "loss": 2.0869, + "step": 94 + }, + { + "epoch": 0.95, + "learning_rate": 3.784932418412385e-05, + "loss": 2.0112, + "step": 95 + }, + { + "epoch": 0.96, + "learning_rate": 3.731856855063486e-05, + "loss": 1.9212, + "step": 96 + }, + { + "epoch": 0.97, + "learning_rate": 3.678649556861724e-05, + "loss": 2.0395, + "step": 97 + }, + { + "epoch": 0.98, + "learning_rate": 3.625325070151155e-05, + "loss": 2.0179, + "step": 98 + }, + { + "epoch": 0.99, + "learning_rate": 3.571897973314004e-05, + "loss": 1.9713, + "step": 99 + }, + { + "epoch": 1.0, + "learning_rate": 3.51838287278508e-05, + "loss": 2.0756, + "step": 100 + }, + { + "epoch": 1.0, + "eval_loss": 2.1577982902526855, + "eval_runtime": 2.9721, + "eval_samples_per_second": 3.365, + "eval_steps_per_second": 0.673, + "step": 100 + }, + { + "epoch": 1.01, + "learning_rate": 3.4647943990585227e-05, + "loss": 1.9634, + "step": 101 + }, + { + "epoch": 1.02, + "learning_rate": 3.411147202687972e-05, + "loss": 2.1004, + "step": 102 + }, + { + "epoch": 1.03, + "learning_rate": 3.357455950281251e-05, + "loss": 1.9162, + "step": 103 + }, + { + "epoch": 1.04, + "learning_rate": 3.303735320490665e-05, + "loss": 1.9874, + "step": 104 + }, + { + "epoch": 1.05, + "learning_rate": 3.25e-05, + "loss": 2.0868, + "step": 105 + }, + { + "epoch": 1.06, + "learning_rate": 3.1962646795093345e-05, + "loss": 1.8818, + "step": 106 + }, + { + "epoch": 1.07, + "learning_rate": 3.142544049718748e-05, + "loss": 1.9699, + "step": 107 + }, + { + "epoch": 1.08, + "learning_rate": 3.088852797312027e-05, + "loss": 1.994, + "step": 108 + }, + { + "epoch": 1.09, + "learning_rate": 3.0352056009414768e-05, + "loss": 2.1169, + "step": 109 + }, + { + "epoch": 1.1, + "learning_rate": 2.98161712721492e-05, + "loss": 1.997, + "step": 110 + }, + { + "epoch": 1.11, + "learning_rate": 2.928102026685996e-05, + "loss": 2.1137, + "step": 111 + }, + { + "epoch": 1.12, + "learning_rate": 2.874674929848845e-05, + "loss": 2.0281, + "step": 112 + }, + { + "epoch": 1.13, + "learning_rate": 2.8213504431382755e-05, + "loss": 2.0649, + "step": 113 + }, + { + "epoch": 1.14, + "learning_rate": 2.768143144936513e-05, + "loss": 1.9891, + "step": 114 + }, + { + "epoch": 1.15, + "learning_rate": 2.715067581587614e-05, + "loss": 1.9841, + "step": 115 + }, + { + "epoch": 1.16, + "learning_rate": 2.6621382634206436e-05, + "loss": 2.1146, + "step": 116 + }, + { + "epoch": 1.17, + "learning_rate": 2.6093696607826892e-05, + "loss": 2.0479, + "step": 117 + }, + { + "epoch": 1.18, + "learning_rate": 2.556776200082823e-05, + "loss": 2.1255, + "step": 118 + }, + { + "epoch": 1.19, + "learning_rate": 2.5043722598480574e-05, + "loss": 2.0523, + "step": 119 + }, + { + "epoch": 1.2, + "learning_rate": 2.4521721667924026e-05, + "loss": 2.1502, + "step": 120 + }, + { + "epoch": 1.2, + "eval_loss": 2.1567764282226562, + "eval_runtime": 2.9942, + "eval_samples_per_second": 3.34, + "eval_steps_per_second": 0.668, + "step": 120 + }, + { + "epoch": 1.21, + "learning_rate": 2.4001901919000855e-05, + "loss": 2.0222, + "step": 121 + }, + { + "epoch": 1.22, + "learning_rate": 2.3484405465240055e-05, + "loss": 2.003, + "step": 122 + }, + { + "epoch": 1.23, + "learning_rate": 2.2969373785004897e-05, + "loss": 1.9935, + "step": 123 + }, + { + "epoch": 1.24, + "learning_rate": 2.2456947682814208e-05, + "loss": 1.92, + "step": 124 + }, + { + "epoch": 1.25, + "learning_rate": 2.1947267250847786e-05, + "loss": 2.0776, + "step": 125 + }, + { + "epoch": 1.26, + "learning_rate": 2.1440471830646597e-05, + "loss": 2.0785, + "step": 126 + }, + { + "epoch": 1.27, + "learning_rate": 2.093669997501821e-05, + "loss": 2.1597, + "step": 127 + }, + { + "epoch": 1.28, + "learning_rate": 2.0436089410157818e-05, + "loss": 2.0279, + "step": 128 + }, + { + "epoch": 1.29, + "learning_rate": 1.993877699799531e-05, + "loss": 2.107, + "step": 129 + }, + { + "epoch": 1.3, + "learning_rate": 1.9444898698778497e-05, + "loss": 2.121, + "step": 130 + }, + { + "epoch": 1.31, + "learning_rate": 1.8954589533903067e-05, + "loss": 2.1024, + "step": 131 + }, + { + "epoch": 1.32, + "learning_rate": 1.8467983548998984e-05, + "loss": 2.1763, + "step": 132 + }, + { + "epoch": 1.33, + "learning_rate": 1.798521377728379e-05, + "loss": 2.0075, + "step": 133 + }, + { + "epoch": 1.34, + "learning_rate": 1.750641220319265e-05, + "loss": 2.0791, + "step": 134 + }, + { + "epoch": 1.35, + "learning_rate": 1.703170972629511e-05, + "loss": 2.1985, + "step": 135 + }, + { + "epoch": 1.36, + "learning_rate": 1.6561236125508495e-05, + "loss": 2.1254, + "step": 136 + }, + { + "epoch": 1.37, + "learning_rate": 1.6095120023617585e-05, + "loss": 2.0442, + "step": 137 + }, + { + "epoch": 1.38, + "learning_rate": 1.5633488852110516e-05, + "loss": 2.0451, + "step": 138 + }, + { + "epoch": 1.39, + "learning_rate": 1.5176468816340209e-05, + "loss": 2.029, + "step": 139 + }, + { + "epoch": 1.4, + "learning_rate": 1.472418486102113e-05, + "loss": 2.108, + "step": 140 + }, + { + "epoch": 1.4, + "eval_loss": 2.153236150741577, + "eval_runtime": 2.9714, + "eval_samples_per_second": 3.365, + "eval_steps_per_second": 0.673, + "step": 140 + }, + { + "epoch": 1.41, + "learning_rate": 1.427676063607064e-05, + "loss": 2.0035, + "step": 141 + }, + { + "epoch": 1.42, + "learning_rate": 1.3834318462804333e-05, + "loss": 2.0776, + "step": 142 + }, + { + "epoch": 1.43, + "learning_rate": 1.3396979300494626e-05, + "loss": 1.9116, + "step": 143 + }, + { + "epoch": 1.44, + "learning_rate": 1.2964862713301684e-05, + "loss": 2.0144, + "step": 144 + }, + { + "epoch": 1.45, + "learning_rate": 1.2538086837585795e-05, + "loss": 2.1723, + "step": 145 + }, + { + "epoch": 1.46, + "learning_rate": 1.2116768349610045e-05, + "loss": 2.0467, + "step": 146 + }, + { + "epoch": 1.47, + "learning_rate": 1.1701022433642207e-05, + "loss": 1.9612, + "step": 147 + }, + { + "epoch": 1.48, + "learning_rate": 1.1290962750464517e-05, + "loss": 2.0506, + "step": 148 + }, + { + "epoch": 1.49, + "learning_rate": 1.0886701406299927e-05, + "loss": 1.9718, + "step": 149 + }, + { + "epoch": 1.5, + "learning_rate": 1.048834892216342e-05, + "loss": 2.0387, + "step": 150 + } + ], + "logging_steps": 1, + "max_steps": 200, + "num_train_epochs": 2, + "save_steps": 50, + "total_flos": 1.251428724071465e+17, + "trial_name": null, + "trial_params": null +} diff --git a/prompt-a/twochar/checkpoint-150/training_args.bin b/prompt-a/twochar/checkpoint-150/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..d58072f53f91b75b135f9498c19f7bcc5b637ba4 --- /dev/null +++ b/prompt-a/twochar/checkpoint-150/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9212c1ed1ebf30dd523c0f488bee8e1daf66e67eaaf76d60029260d57f02f88 +size 4411 diff --git a/prompt-a/twochar/checkpoint-200/README.md b/prompt-a/twochar/checkpoint-200/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/prompt-a/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/prompt-a/twochar/checkpoint-200/adapter_config.json b/prompt-a/twochar/checkpoint-200/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..e15b2402bceaab5939e21de3eb1d9a6ad2980bbd --- /dev/null +++ b/prompt-a/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": [ + "down_proj", + "k_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/prompt-a/twochar/checkpoint-200/adapter_model.bin b/prompt-a/twochar/checkpoint-200/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..c96c513e5c63ebcef49c385fe35a3a5331b09141 --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3686a6854a70ed7424cb6434eaf2d2d208980a99034b0d2075128de53b3297a2 +size 125374989 diff --git a/prompt-a/twochar/checkpoint-200/adapter_model/README.md b/prompt-a/twochar/checkpoint-200/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/prompt-a/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/prompt-a/twochar/checkpoint-200/adapter_model/adapter_config.json b/prompt-a/twochar/checkpoint-200/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..e15b2402bceaab5939e21de3eb1d9a6ad2980bbd --- /dev/null +++ b/prompt-a/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": [ + "down_proj", + "k_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/prompt-a/twochar/checkpoint-200/adapter_model/adapter_model.bin b/prompt-a/twochar/checkpoint-200/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..c96c513e5c63ebcef49c385fe35a3a5331b09141 --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3686a6854a70ed7424cb6434eaf2d2d208980a99034b0d2075128de53b3297a2 +size 125374989 diff --git a/prompt-a/twochar/checkpoint-200/optimizer.pt b/prompt-a/twochar/checkpoint-200/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..b91c9b3a87116ba9c5b51128f99efb6b48580d97 --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:85cc40f0d7f81e4b3136a22c9ee61293c7e2a46795fded91d441e599b632fe7b +size 250681597 diff --git a/prompt-a/twochar/checkpoint-200/rng_state_0.pth b/prompt-a/twochar/checkpoint-200/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..0b6fb0a77ef10e58c647935eb9df2dc9d1b48312 --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cb71352271decccd21025740fb7c4585441f353b8cbc20f09b524c2fe7cf027d +size 21687 diff --git a/prompt-a/twochar/checkpoint-200/rng_state_1.pth b/prompt-a/twochar/checkpoint-200/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..ecb36eaa5ead1a2f52fd2d4368407b6efddff9d7 --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bd2535ba0a236e9acf8608fce0f4038a69c182287799399a90be854fb25b7c34 +size 21687 diff --git a/prompt-a/twochar/checkpoint-200/rng_state_2.pth b/prompt-a/twochar/checkpoint-200/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..9324b5ad6d9e39fa976dfa1f629f78807c7ae59a --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eba2c62c4b80d00ace4ba0acc8ebdad176eddf3cd3a5b1050f7df7c4700163a6 +size 21687 diff --git a/prompt-a/twochar/checkpoint-200/rng_state_3.pth b/prompt-a/twochar/checkpoint-200/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..569379c20034aed4344efb81bf440feccffe600e --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cb2fcfaf47b0c610c3711de92cc4604e5d6b41802b02ec9b5c582f4f729af8f2 +size 21687 diff --git a/prompt-a/twochar/checkpoint-200/rng_state_4.pth b/prompt-a/twochar/checkpoint-200/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..16fe71ba3916826371f2252eb0fbc2e7ea31f0dc --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1a3d6d3323887d5d66971f041a51221a123e538e2a897b4864257eadd27473e1 +size 21687 diff --git a/prompt-a/twochar/checkpoint-200/rng_state_5.pth b/prompt-a/twochar/checkpoint-200/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..0da27d0b1c857e40f507b34e3e8e50c680935545 --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b1fb79dd2434c5cd9a6946f1fb7f838110f92f121c85810b9185b6e281a9b0b2 +size 21687 diff --git a/prompt-a/twochar/checkpoint-200/rng_state_6.pth b/prompt-a/twochar/checkpoint-200/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..e9ff0b3f4b969a683a222a53eff3f29de16f65d8 --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:458ac70464bdf3b43495369875d77f8490d72156a48c1118ce85ad717e807a7b +size 21687 diff --git a/prompt-a/twochar/checkpoint-200/rng_state_7.pth b/prompt-a/twochar/checkpoint-200/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..62e7103eef41d1525ab0ca9c0819f4477a68338c --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:198429cb8ecd769bf49f460c1af624920d6cbc0343641d33de69f2dd82bca33f +size 21687 diff --git a/prompt-a/twochar/checkpoint-200/scheduler.pt b/prompt-a/twochar/checkpoint-200/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..deea23988d426e71f3857d834f7bd88d8ce96a10 --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6afe3ae57ca7ad0c3a1c16a19bcce025d43cf50175f27f254174a0652c2ad2b3 +size 627 diff --git a/prompt-a/twochar/checkpoint-200/trainer_state.json b/prompt-a/twochar/checkpoint-200/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..932b2f2e443e56057492d3f9e316190ab4f16b42 --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/trainer_state.json @@ -0,0 +1,1299 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.0, + "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": 1.9562, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 1.3e-05, + "loss": 2.0273, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 1.9499999999999996e-05, + "loss": 2.2516, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 2.6e-05, + "loss": 2.1955, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 3.25e-05, + "loss": 2.1752, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 3.899999999999999e-05, + "loss": 2.2144, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 4.5499999999999995e-05, + "loss": 2.0829, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 5.2e-05, + "loss": 2.0503, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 5.85e-05, + "loss": 2.1558, + "step": 9 + }, + { + "epoch": 0.1, + "learning_rate": 6.5e-05, + "loss": 2.0924, + "step": 10 + }, + { + "epoch": 0.11, + "learning_rate": 6.499555741225524e-05, + "loss": 2.1238, + "step": 11 + }, + { + "epoch": 0.12, + "learning_rate": 6.498223086358009e-05, + "loss": 2.129, + "step": 12 + }, + { + "epoch": 0.13, + "learning_rate": 6.49600239973199e-05, + "loss": 2.0559, + "step": 13 + }, + { + "epoch": 0.14, + "learning_rate": 6.492894288461016e-05, + "loss": 2.0659, + "step": 14 + }, + { + "epoch": 0.15, + "learning_rate": 6.488899602271677e-05, + "loss": 2.1563, + "step": 15 + }, + { + "epoch": 0.16, + "learning_rate": 6.484019433271287e-05, + "loss": 2.1706, + "step": 16 + }, + { + "epoch": 0.17, + "learning_rate": 6.478255115649324e-05, + "loss": 2.0736, + "step": 17 + }, + { + "epoch": 0.18, + "learning_rate": 6.471608225312668e-05, + "loss": 2.1295, + "step": 18 + }, + { + "epoch": 0.19, + "learning_rate": 6.464080579454772e-05, + "loss": 2.0618, + "step": 19 + }, + { + "epoch": 0.2, + "learning_rate": 6.455674236058847e-05, + "loss": 2.12, + "step": 20 + }, + { + "epoch": 0.2, + "eval_loss": 2.1955978870391846, + "eval_runtime": 2.9627, + "eval_samples_per_second": 3.375, + "eval_steps_per_second": 0.675, + "step": 20 + }, + { + "epoch": 0.21, + "learning_rate": 6.446391493335242e-05, + "loss": 2.1576, + "step": 21 + }, + { + "epoch": 0.22, + "learning_rate": 6.43623488909313e-05, + "loss": 2.0446, + "step": 22 + }, + { + "epoch": 0.23, + "learning_rate": 6.425207200046697e-05, + "loss": 2.0033, + "step": 23 + }, + { + "epoch": 0.24, + "learning_rate": 6.413311441056019e-05, + "loss": 2.1382, + "step": 24 + }, + { + "epoch": 0.25, + "learning_rate": 6.400550864302823e-05, + "loss": 2.0989, + "step": 25 + }, + { + "epoch": 0.26, + "learning_rate": 6.386928958401383e-05, + "loss": 2.1276, + "step": 26 + }, + { + "epoch": 0.27, + "learning_rate": 6.372449447444757e-05, + "loss": 2.023, + "step": 27 + }, + { + "epoch": 0.28, + "learning_rate": 6.357116289986662e-05, + "loss": 2.0597, + "step": 28 + }, + { + "epoch": 0.29, + "learning_rate": 6.340933677959248e-05, + "loss": 2.0529, + "step": 29 + }, + { + "epoch": 0.3, + "learning_rate": 6.323906035527062e-05, + "loss": 1.9873, + "step": 30 + }, + { + "epoch": 0.31, + "learning_rate": 6.306038017877523e-05, + "loss": 2.0613, + "step": 31 + }, + { + "epoch": 0.32, + "learning_rate": 6.287334509948247e-05, + "loss": 2.0765, + "step": 32 + }, + { + "epoch": 0.33, + "learning_rate": 6.267800625091547e-05, + "loss": 2.0809, + "step": 33 + }, + { + "epoch": 0.34, + "learning_rate": 6.2474417036765e-05, + "loss": 2.072, + "step": 34 + }, + { + "epoch": 0.35, + "learning_rate": 6.226263311628936e-05, + "loss": 2.0069, + "step": 35 + }, + { + "epoch": 0.36, + "learning_rate": 6.204271238909777e-05, + "loss": 2.0784, + "step": 36 + }, + { + "epoch": 0.37, + "learning_rate": 6.181471497932117e-05, + "loss": 2.1763, + "step": 37 + }, + { + "epoch": 0.38, + "learning_rate": 6.157870321917481e-05, + "loss": 2.0402, + "step": 38 + }, + { + "epoch": 0.39, + "learning_rate": 6.133474163191738e-05, + "loss": 2.1082, + "step": 39 + }, + { + "epoch": 0.4, + "learning_rate": 6.108289691421088e-05, + "loss": 2.1381, + "step": 40 + }, + { + "epoch": 0.4, + "eval_loss": 2.172107219696045, + "eval_runtime": 2.97, + "eval_samples_per_second": 3.367, + "eval_steps_per_second": 0.673, + "step": 40 + }, + { + "epoch": 0.41, + "learning_rate": 6.08232379178865e-05, + "loss": 2.0538, + "step": 41 + }, + { + "epoch": 0.42, + "learning_rate": 6.055583563112115e-05, + "loss": 2.222, + "step": 42 + }, + { + "epoch": 0.43, + "learning_rate": 6.0280763159030015e-05, + "loss": 1.996, + "step": 43 + }, + { + "epoch": 0.44, + "learning_rate": 5.999809570368039e-05, + "loss": 1.9798, + "step": 44 + }, + { + "epoch": 0.45, + "learning_rate": 5.970791054353217e-05, + "loss": 2.1825, + "step": 45 + }, + { + "epoch": 0.46, + "learning_rate": 5.941028701231065e-05, + "loss": 2.1671, + "step": 46 + }, + { + "epoch": 0.47, + "learning_rate": 5.91053064773175e-05, + "loss": 2.0992, + "step": 47 + }, + { + "epoch": 0.48, + "learning_rate": 5.8793052317185785e-05, + "loss": 2.0544, + "step": 48 + }, + { + "epoch": 0.49, + "learning_rate": 5.8473609899085046e-05, + "loss": 2.1054, + "step": 49 + }, + { + "epoch": 0.5, + "learning_rate": 5.814706655538279e-05, + "loss": 2.1214, + "step": 50 + }, + { + "epoch": 0.51, + "learning_rate": 5.781351155976865e-05, + "loss": 1.9961, + "step": 51 + }, + { + "epoch": 0.52, + "learning_rate": 5.7473036102847915e-05, + "loss": 2.0233, + "step": 52 + }, + { + "epoch": 0.53, + "learning_rate": 5.712573326721088e-05, + "loss": 2.0756, + "step": 53 + }, + { + "epoch": 0.54, + "learning_rate": 5.6771698001984996e-05, + "loss": 2.0138, + "step": 54 + }, + { + "epoch": 0.55, + "learning_rate": 5.641102709687678e-05, + "loss": 1.9964, + "step": 55 + }, + { + "epoch": 0.56, + "learning_rate": 5.604381915571039e-05, + "loss": 2.0915, + "step": 56 + }, + { + "epoch": 0.57, + "learning_rate": 5.5670174569470405e-05, + "loss": 1.9954, + "step": 57 + }, + { + "epoch": 0.58, + "learning_rate": 5.529019548885587e-05, + "loss": 2.1395, + "step": 58 + }, + { + "epoch": 0.59, + "learning_rate": 5.490398579635338e-05, + "loss": 2.0388, + "step": 59 + }, + { + "epoch": 0.6, + "learning_rate": 5.451165107783658e-05, + "loss": 2.0979, + "step": 60 + }, + { + "epoch": 0.6, + "eval_loss": 2.166550874710083, + "eval_runtime": 2.9777, + "eval_samples_per_second": 3.358, + "eval_steps_per_second": 0.672, + "step": 60 + }, + { + "epoch": 0.61, + "learning_rate": 5.411329859370006e-05, + "loss": 2.0614, + "step": 61 + }, + { + "epoch": 0.62, + "learning_rate": 5.370903724953548e-05, + "loss": 2.061, + "step": 62 + }, + { + "epoch": 0.63, + "learning_rate": 5.329897756635779e-05, + "loss": 2.0868, + "step": 63 + }, + { + "epoch": 0.64, + "learning_rate": 5.2883231650389946e-05, + "loss": 2.0659, + "step": 64 + }, + { + "epoch": 0.65, + "learning_rate": 5.2461913162414206e-05, + "loss": 2.1325, + "step": 65 + }, + { + "epoch": 0.66, + "learning_rate": 5.203513728669831e-05, + "loss": 2.0955, + "step": 66 + }, + { + "epoch": 0.67, + "learning_rate": 5.160302069950537e-05, + "loss": 1.9692, + "step": 67 + }, + { + "epoch": 0.68, + "learning_rate": 5.116568153719566e-05, + "loss": 2.0858, + "step": 68 + }, + { + "epoch": 0.69, + "learning_rate": 5.072323936392936e-05, + "loss": 2.0884, + "step": 69 + }, + { + "epoch": 0.7, + "learning_rate": 5.027581513897888e-05, + "loss": 1.952, + "step": 70 + }, + { + "epoch": 0.71, + "learning_rate": 4.9823531183659796e-05, + "loss": 2.0223, + "step": 71 + }, + { + "epoch": 0.72, + "learning_rate": 4.936651114788949e-05, + "loss": 2.0098, + "step": 72 + }, + { + "epoch": 0.73, + "learning_rate": 4.890487997638242e-05, + "loss": 1.9965, + "step": 73 + }, + { + "epoch": 0.74, + "learning_rate": 4.843876387449151e-05, + "loss": 2.1097, + "step": 74 + }, + { + "epoch": 0.75, + "learning_rate": 4.796829027370489e-05, + "loss": 2.1051, + "step": 75 + }, + { + "epoch": 0.76, + "learning_rate": 4.749358779680735e-05, + "loss": 2.1672, + "step": 76 + }, + { + "epoch": 0.77, + "learning_rate": 4.701478622271621e-05, + "loss": 2.0641, + "step": 77 + }, + { + "epoch": 0.78, + "learning_rate": 4.653201645100101e-05, + "loss": 2.1899, + "step": 78 + }, + { + "epoch": 0.79, + "learning_rate": 4.6045410466096934e-05, + "loss": 2.007, + "step": 79 + }, + { + "epoch": 0.8, + "learning_rate": 4.555510130122151e-05, + "loss": 2.1355, + "step": 80 + }, + { + "epoch": 0.8, + "eval_loss": 2.160327434539795, + "eval_runtime": 2.9723, + "eval_samples_per_second": 3.364, + "eval_steps_per_second": 0.673, + "step": 80 + }, + { + "epoch": 0.81, + "learning_rate": 4.5061223002004694e-05, + "loss": 2.0046, + "step": 81 + }, + { + "epoch": 0.82, + "learning_rate": 4.456391058984217e-05, + "loss": 2.0902, + "step": 82 + }, + { + "epoch": 0.83, + "learning_rate": 4.406330002498178e-05, + "loss": 2.1259, + "step": 83 + }, + { + "epoch": 0.84, + "learning_rate": 4.3559528169353394e-05, + "loss": 1.9241, + "step": 84 + }, + { + "epoch": 0.85, + "learning_rate": 4.3052732749152215e-05, + "loss": 2.1537, + "step": 85 + }, + { + "epoch": 0.86, + "learning_rate": 4.254305231718579e-05, + "loss": 1.9482, + "step": 86 + }, + { + "epoch": 0.87, + "learning_rate": 4.2030626214995104e-05, + "loss": 2.0181, + "step": 87 + }, + { + "epoch": 0.88, + "learning_rate": 4.151559453475994e-05, + "loss": 1.9839, + "step": 88 + }, + { + "epoch": 0.89, + "learning_rate": 4.0998098080999136e-05, + "loss": 2.0735, + "step": 89 + }, + { + "epoch": 0.9, + "learning_rate": 4.047827833207597e-05, + "loss": 2.1691, + "step": 90 + }, + { + "epoch": 0.91, + "learning_rate": 3.995627740151942e-05, + "loss": 2.2761, + "step": 91 + }, + { + "epoch": 0.92, + "learning_rate": 3.9432237999171774e-05, + "loss": 2.1657, + "step": 92 + }, + { + "epoch": 0.93, + "learning_rate": 3.89063033921731e-05, + "loss": 1.9501, + "step": 93 + }, + { + "epoch": 0.94, + "learning_rate": 3.8378617365793576e-05, + "loss": 2.0869, + "step": 94 + }, + { + "epoch": 0.95, + "learning_rate": 3.784932418412385e-05, + "loss": 2.0112, + "step": 95 + }, + { + "epoch": 0.96, + "learning_rate": 3.731856855063486e-05, + "loss": 1.9212, + "step": 96 + }, + { + "epoch": 0.97, + "learning_rate": 3.678649556861724e-05, + "loss": 2.0395, + "step": 97 + }, + { + "epoch": 0.98, + "learning_rate": 3.625325070151155e-05, + "loss": 2.0179, + "step": 98 + }, + { + "epoch": 0.99, + "learning_rate": 3.571897973314004e-05, + "loss": 1.9713, + "step": 99 + }, + { + "epoch": 1.0, + "learning_rate": 3.51838287278508e-05, + "loss": 2.0756, + "step": 100 + }, + { + "epoch": 1.0, + "eval_loss": 2.1577982902526855, + "eval_runtime": 2.9721, + "eval_samples_per_second": 3.365, + "eval_steps_per_second": 0.673, + "step": 100 + }, + { + "epoch": 1.01, + "learning_rate": 3.4647943990585227e-05, + "loss": 1.9634, + "step": 101 + }, + { + "epoch": 1.02, + "learning_rate": 3.411147202687972e-05, + "loss": 2.1004, + "step": 102 + }, + { + "epoch": 1.03, + "learning_rate": 3.357455950281251e-05, + "loss": 1.9162, + "step": 103 + }, + { + "epoch": 1.04, + "learning_rate": 3.303735320490665e-05, + "loss": 1.9874, + "step": 104 + }, + { + "epoch": 1.05, + "learning_rate": 3.25e-05, + "loss": 2.0868, + "step": 105 + }, + { + "epoch": 1.06, + "learning_rate": 3.1962646795093345e-05, + "loss": 1.8818, + "step": 106 + }, + { + "epoch": 1.07, + "learning_rate": 3.142544049718748e-05, + "loss": 1.9699, + "step": 107 + }, + { + "epoch": 1.08, + "learning_rate": 3.088852797312027e-05, + "loss": 1.994, + "step": 108 + }, + { + "epoch": 1.09, + "learning_rate": 3.0352056009414768e-05, + "loss": 2.1169, + "step": 109 + }, + { + "epoch": 1.1, + "learning_rate": 2.98161712721492e-05, + "loss": 1.997, + "step": 110 + }, + { + "epoch": 1.11, + "learning_rate": 2.928102026685996e-05, + "loss": 2.1137, + "step": 111 + }, + { + "epoch": 1.12, + "learning_rate": 2.874674929848845e-05, + "loss": 2.0281, + "step": 112 + }, + { + "epoch": 1.13, + "learning_rate": 2.8213504431382755e-05, + "loss": 2.0649, + "step": 113 + }, + { + "epoch": 1.14, + "learning_rate": 2.768143144936513e-05, + "loss": 1.9891, + "step": 114 + }, + { + "epoch": 1.15, + "learning_rate": 2.715067581587614e-05, + "loss": 1.9841, + "step": 115 + }, + { + "epoch": 1.16, + "learning_rate": 2.6621382634206436e-05, + "loss": 2.1146, + "step": 116 + }, + { + "epoch": 1.17, + "learning_rate": 2.6093696607826892e-05, + "loss": 2.0479, + "step": 117 + }, + { + "epoch": 1.18, + "learning_rate": 2.556776200082823e-05, + "loss": 2.1255, + "step": 118 + }, + { + "epoch": 1.19, + "learning_rate": 2.5043722598480574e-05, + "loss": 2.0523, + "step": 119 + }, + { + "epoch": 1.2, + "learning_rate": 2.4521721667924026e-05, + "loss": 2.1502, + "step": 120 + }, + { + "epoch": 1.2, + "eval_loss": 2.1567764282226562, + "eval_runtime": 2.9942, + "eval_samples_per_second": 3.34, + "eval_steps_per_second": 0.668, + "step": 120 + }, + { + "epoch": 1.21, + "learning_rate": 2.4001901919000855e-05, + "loss": 2.0222, + "step": 121 + }, + { + "epoch": 1.22, + "learning_rate": 2.3484405465240055e-05, + "loss": 2.003, + "step": 122 + }, + { + "epoch": 1.23, + "learning_rate": 2.2969373785004897e-05, + "loss": 1.9935, + "step": 123 + }, + { + "epoch": 1.24, + "learning_rate": 2.2456947682814208e-05, + "loss": 1.92, + "step": 124 + }, + { + "epoch": 1.25, + "learning_rate": 2.1947267250847786e-05, + "loss": 2.0776, + "step": 125 + }, + { + "epoch": 1.26, + "learning_rate": 2.1440471830646597e-05, + "loss": 2.0785, + "step": 126 + }, + { + "epoch": 1.27, + "learning_rate": 2.093669997501821e-05, + "loss": 2.1597, + "step": 127 + }, + { + "epoch": 1.28, + "learning_rate": 2.0436089410157818e-05, + "loss": 2.0279, + "step": 128 + }, + { + "epoch": 1.29, + "learning_rate": 1.993877699799531e-05, + "loss": 2.107, + "step": 129 + }, + { + "epoch": 1.3, + "learning_rate": 1.9444898698778497e-05, + "loss": 2.121, + "step": 130 + }, + { + "epoch": 1.31, + "learning_rate": 1.8954589533903067e-05, + "loss": 2.1024, + "step": 131 + }, + { + "epoch": 1.32, + "learning_rate": 1.8467983548998984e-05, + "loss": 2.1763, + "step": 132 + }, + { + "epoch": 1.33, + "learning_rate": 1.798521377728379e-05, + "loss": 2.0075, + "step": 133 + }, + { + "epoch": 1.34, + "learning_rate": 1.750641220319265e-05, + "loss": 2.0791, + "step": 134 + }, + { + "epoch": 1.35, + "learning_rate": 1.703170972629511e-05, + "loss": 2.1985, + "step": 135 + }, + { + "epoch": 1.36, + "learning_rate": 1.6561236125508495e-05, + "loss": 2.1254, + "step": 136 + }, + { + "epoch": 1.37, + "learning_rate": 1.6095120023617585e-05, + "loss": 2.0442, + "step": 137 + }, + { + "epoch": 1.38, + "learning_rate": 1.5633488852110516e-05, + "loss": 2.0451, + "step": 138 + }, + { + "epoch": 1.39, + "learning_rate": 1.5176468816340209e-05, + "loss": 2.029, + "step": 139 + }, + { + "epoch": 1.4, + "learning_rate": 1.472418486102113e-05, + "loss": 2.108, + "step": 140 + }, + { + "epoch": 1.4, + "eval_loss": 2.153236150741577, + "eval_runtime": 2.9714, + "eval_samples_per_second": 3.365, + "eval_steps_per_second": 0.673, + "step": 140 + }, + { + "epoch": 1.41, + "learning_rate": 1.427676063607064e-05, + "loss": 2.0035, + "step": 141 + }, + { + "epoch": 1.42, + "learning_rate": 1.3834318462804333e-05, + "loss": 2.0776, + "step": 142 + }, + { + "epoch": 1.43, + "learning_rate": 1.3396979300494626e-05, + "loss": 1.9116, + "step": 143 + }, + { + "epoch": 1.44, + "learning_rate": 1.2964862713301684e-05, + "loss": 2.0144, + "step": 144 + }, + { + "epoch": 1.45, + "learning_rate": 1.2538086837585795e-05, + "loss": 2.1723, + "step": 145 + }, + { + "epoch": 1.46, + "learning_rate": 1.2116768349610045e-05, + "loss": 2.0467, + "step": 146 + }, + { + "epoch": 1.47, + "learning_rate": 1.1701022433642207e-05, + "loss": 1.9612, + "step": 147 + }, + { + "epoch": 1.48, + "learning_rate": 1.1290962750464517e-05, + "loss": 2.0506, + "step": 148 + }, + { + "epoch": 1.49, + "learning_rate": 1.0886701406299927e-05, + "loss": 1.9718, + "step": 149 + }, + { + "epoch": 1.5, + "learning_rate": 1.048834892216342e-05, + "loss": 2.0387, + "step": 150 + }, + { + "epoch": 1.51, + "learning_rate": 1.009601420364661e-05, + "loss": 1.9916, + "step": 151 + }, + { + "epoch": 1.52, + "learning_rate": 9.70980451114412e-06, + "loss": 2.1382, + "step": 152 + }, + { + "epoch": 1.53, + "learning_rate": 9.329825430529596e-06, + "loss": 1.8168, + "step": 153 + }, + { + "epoch": 1.54, + "learning_rate": 8.956180844289608e-06, + "loss": 2.0349, + "step": 154 + }, + { + "epoch": 1.55, + "learning_rate": 8.588972903123222e-06, + "loss": 2.1169, + "step": 155 + }, + { + "epoch": 1.56, + "learning_rate": 8.228301998015008e-06, + "loss": 2.0028, + "step": 156 + }, + { + "epoch": 1.57, + "learning_rate": 7.874266732789127e-06, + "loss": 1.9355, + "step": 157 + }, + { + "epoch": 1.58, + "learning_rate": 7.526963897152088e-06, + "loss": 2.117, + "step": 158 + }, + { + "epoch": 1.59, + "learning_rate": 7.1864884402313486e-06, + "loss": 1.8699, + "step": 159 + }, + { + "epoch": 1.6, + "learning_rate": 6.8529334446172116e-06, + "loss": 2.0929, + "step": 160 + }, + { + "epoch": 1.6, + "eval_loss": 2.152268409729004, + "eval_runtime": 2.971, + "eval_samples_per_second": 3.366, + "eval_steps_per_second": 0.673, + "step": 160 + }, + { + "epoch": 1.61, + "learning_rate": 6.526390100914946e-06, + "loss": 2.002, + "step": 161 + }, + { + "epoch": 1.62, + "learning_rate": 6.206947682814211e-06, + "loss": 2.0595, + "step": 162 + }, + { + "epoch": 1.63, + "learning_rate": 5.894693522682498e-06, + "loss": 2.1649, + "step": 163 + }, + { + "epoch": 1.64, + "learning_rate": 5.589712987689348e-06, + "loss": 2.0812, + "step": 164 + }, + { + "epoch": 1.65, + "learning_rate": 5.292089456467822e-06, + "loss": 2.0669, + "step": 165 + }, + { + "epoch": 1.66, + "learning_rate": 5.001904296319595e-06, + "loss": 1.9335, + "step": 166 + }, + { + "epoch": 1.67, + "learning_rate": 4.7192368409699775e-06, + "loss": 2.0828, + "step": 167 + }, + { + "epoch": 1.68, + "learning_rate": 4.444164368878848e-06, + "loss": 2.0618, + "step": 168 + }, + { + "epoch": 1.69, + "learning_rate": 4.1767620821134845e-06, + "loss": 1.9768, + "step": 169 + }, + { + "epoch": 1.7, + "learning_rate": 3.917103085789108e-06, + "loss": 2.0585, + "step": 170 + }, + { + "epoch": 1.71, + "learning_rate": 3.665258368082616e-06, + "loss": 2.2143, + "step": 171 + }, + { + "epoch": 1.72, + "learning_rate": 3.4212967808251866e-06, + "loss": 1.9548, + "step": 172 + }, + { + "epoch": 1.73, + "learning_rate": 3.1852850206788356e-06, + "loss": 1.9494, + "step": 173 + }, + { + "epoch": 1.74, + "learning_rate": 2.957287610902218e-06, + "loss": 1.9549, + "step": 174 + }, + { + "epoch": 1.75, + "learning_rate": 2.7373668837106307e-06, + "loss": 2.0845, + "step": 175 + }, + { + "epoch": 1.76, + "learning_rate": 2.5255829632349955e-06, + "loss": 2.0526, + "step": 176 + }, + { + "epoch": 1.77, + "learning_rate": 2.3219937490845233e-06, + "loss": 2.0889, + "step": 177 + }, + { + "epoch": 1.78, + "learning_rate": 2.1266549005175255e-06, + "loss": 2.0376, + "step": 178 + }, + { + "epoch": 1.79, + "learning_rate": 1.9396198212247617e-06, + "loss": 1.9105, + "step": 179 + }, + { + "epoch": 1.8, + "learning_rate": 1.7609396447293741e-06, + "loss": 1.8756, + "step": 180 + }, + { + "epoch": 1.8, + "eval_loss": 2.1519551277160645, + "eval_runtime": 2.9671, + "eval_samples_per_second": 3.37, + "eval_steps_per_second": 0.674, + "step": 180 + }, + { + "epoch": 1.81, + "learning_rate": 1.5906632204075102e-06, + "loss": 2.0699, + "step": 181 + }, + { + "epoch": 1.82, + "learning_rate": 1.4288371001333764e-06, + "loss": 2.1844, + "step": 182 + }, + { + "epoch": 1.83, + "learning_rate": 1.2755055255524352e-06, + "loss": 1.8558, + "step": 183 + }, + { + "epoch": 1.84, + "learning_rate": 1.1307104159861715e-06, + "loss": 2.0213, + "step": 184 + }, + { + "epoch": 1.85, + "learning_rate": 9.944913569717627e-07, + "loss": 1.9312, + "step": 185 + }, + { + "epoch": 1.86, + "learning_rate": 8.668855894398133e-07, + "loss": 1.9671, + "step": 186 + }, + { + "epoch": 1.87, + "learning_rate": 7.479279995330236e-07, + "loss": 1.9802, + "step": 187 + }, + { + "epoch": 1.88, + "learning_rate": 6.376511090686959e-07, + "loss": 2.0252, + "step": 188 + }, + { + "epoch": 1.89, + "learning_rate": 5.36085066647573e-07, + "loss": 2.0668, + "step": 189 + }, + { + "epoch": 1.9, + "learning_rate": 4.4325763941152447e-07, + "loss": 1.9095, + "step": 190 + }, + { + "epoch": 1.91, + "learning_rate": 3.5919420545227397e-07, + "loss": 2.0075, + "step": 191 + }, + { + "epoch": 1.92, + "learning_rate": 2.8391774687330616e-07, + "loss": 2.0847, + "step": 192 + }, + { + "epoch": 1.93, + "learning_rate": 2.1744884350676035e-07, + "loss": 1.9515, + "step": 193 + }, + { + "epoch": 1.94, + "learning_rate": 1.5980566728712895e-07, + "loss": 2.0127, + "step": 194 + }, + { + "epoch": 1.95, + "learning_rate": 1.1100397728322997e-07, + "loss": 1.9781, + "step": 195 + }, + { + "epoch": 1.96, + "learning_rate": 7.105711538982761e-08, + "loss": 1.8445, + "step": 196 + }, + { + "epoch": 1.97, + "learning_rate": 3.997600268009604e-08, + "loss": 1.905, + "step": 197 + }, + { + "epoch": 1.98, + "learning_rate": 1.7769136419903407e-08, + "loss": 1.994, + "step": 198 + }, + { + "epoch": 1.99, + "learning_rate": 4.442587744757365e-09, + "loss": 2.0697, + "step": 199 + }, + { + "epoch": 2.0, + "learning_rate": 0.0, + "loss": 2.0917, + "step": 200 + }, + { + "epoch": 2.0, + "eval_loss": 2.1522109508514404, + "eval_runtime": 2.9686, + "eval_samples_per_second": 3.369, + "eval_steps_per_second": 0.674, + "step": 200 + } + ], + "logging_steps": 1, + "max_steps": 200, + "num_train_epochs": 2, + "save_steps": 50, + "total_flos": 1.6723216094252237e+17, + "trial_name": null, + "trial_params": null +} diff --git a/prompt-a/twochar/checkpoint-200/training_args.bin b/prompt-a/twochar/checkpoint-200/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..d58072f53f91b75b135f9498c19f7bcc5b637ba4 --- /dev/null +++ b/prompt-a/twochar/checkpoint-200/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9212c1ed1ebf30dd523c0f488bee8e1daf66e67eaaf76d60029260d57f02f88 +size 4411 diff --git a/prompt-a/twochar/checkpoint-50/README.md b/prompt-a/twochar/checkpoint-50/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/prompt-a/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/prompt-a/twochar/checkpoint-50/adapter_config.json b/prompt-a/twochar/checkpoint-50/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..e15b2402bceaab5939e21de3eb1d9a6ad2980bbd --- /dev/null +++ b/prompt-a/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": [ + "down_proj", + "k_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/prompt-a/twochar/checkpoint-50/adapter_model.bin b/prompt-a/twochar/checkpoint-50/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..b32d692e503b2fe7d2460d16d257316e977bc030 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d840bb206b581a6ed50a4e2191a12b71cc79e1cbc78a648fb9168f1c00985f5d +size 125374989 diff --git a/prompt-a/twochar/checkpoint-50/adapter_model/README.md b/prompt-a/twochar/checkpoint-50/adapter_model/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1e3637f645b79c1dff559d466047b102e3892f5d --- /dev/null +++ b/prompt-a/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/prompt-a/twochar/checkpoint-50/adapter_model/adapter_config.json b/prompt-a/twochar/checkpoint-50/adapter_model/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..e15b2402bceaab5939e21de3eb1d9a6ad2980bbd --- /dev/null +++ b/prompt-a/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": [ + "down_proj", + "k_proj", + "o_proj", + "v_proj", + "up_proj", + "q_proj", + "gate_proj" + ], + "task_type": "CAUSAL_LM" +} \ No newline at end of file diff --git a/prompt-a/twochar/checkpoint-50/adapter_model/adapter_model.bin b/prompt-a/twochar/checkpoint-50/adapter_model/adapter_model.bin new file mode 100644 index 0000000000000000000000000000000000000000..b32d692e503b2fe7d2460d16d257316e977bc030 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/adapter_model/adapter_model.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d840bb206b581a6ed50a4e2191a12b71cc79e1cbc78a648fb9168f1c00985f5d +size 125374989 diff --git a/prompt-a/twochar/checkpoint-50/optimizer.pt b/prompt-a/twochar/checkpoint-50/optimizer.pt new file mode 100644 index 0000000000000000000000000000000000000000..49ac3f1cc9a2674ebb439613a7e39777890234c2 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/optimizer.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:473dc7891d944ed624fd40da875a8426b5821d4124ce68859d629ae31a043331 +size 250681597 diff --git a/prompt-a/twochar/checkpoint-50/rng_state_0.pth b/prompt-a/twochar/checkpoint-50/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..5eb3a4edab5539e7094148dfe79adbe66f009b23 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3a8f71da32d999311dd9a36a9c6572fd4ca746e4452d5398c8f8c76bfb2688d6 +size 21687 diff --git a/prompt-a/twochar/checkpoint-50/rng_state_1.pth b/prompt-a/twochar/checkpoint-50/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..b3073ead5c80b18ad7fd8314f1928cfc50ab4add --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ba2c07e342aa482f28d55882c49641acf46cc55125b6b51531e6e3f2bcacaeae +size 21687 diff --git a/prompt-a/twochar/checkpoint-50/rng_state_2.pth b/prompt-a/twochar/checkpoint-50/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..28afca9c021658b97f4e5b445407275eec76e1a1 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:11d43713e09c4fc5cb64856b182d04e6b740881eab6f1f0534db9efbe16932e5 +size 21687 diff --git a/prompt-a/twochar/checkpoint-50/rng_state_3.pth b/prompt-a/twochar/checkpoint-50/rng_state_3.pth new file mode 100644 index 0000000000000000000000000000000000000000..f0e6ca75d45273e6ebc7acb8687bccffb7976224 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/rng_state_3.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b9e340ac3a586f35d54d09bd32d1a46c322776319814c13e0eacff9eeac97996 +size 21687 diff --git a/prompt-a/twochar/checkpoint-50/rng_state_4.pth b/prompt-a/twochar/checkpoint-50/rng_state_4.pth new file mode 100644 index 0000000000000000000000000000000000000000..153694a41821b7130dbc2a87696a50d7ac275827 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/rng_state_4.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:db165393406425492c97925ce8b799592ff36152d3326fc7b571db8b86a48482 +size 21687 diff --git a/prompt-a/twochar/checkpoint-50/rng_state_5.pth b/prompt-a/twochar/checkpoint-50/rng_state_5.pth new file mode 100644 index 0000000000000000000000000000000000000000..ea5efb06eff095b8ed0792688131a2e36981e3a9 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/rng_state_5.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:13196792301f7e45f8d68c4b33d57ccd0e52c5b2bb79af3bfec6cd2d1c7b3e9d +size 21687 diff --git a/prompt-a/twochar/checkpoint-50/rng_state_6.pth b/prompt-a/twochar/checkpoint-50/rng_state_6.pth new file mode 100644 index 0000000000000000000000000000000000000000..ecdc9981c4bcc00853b7038b96a7d52ecfede8b1 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/rng_state_6.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:016ebea38ff490b7c931a7e76a166feca870ac0396e5d93c15dbb6884dc8029f +size 21687 diff --git a/prompt-a/twochar/checkpoint-50/rng_state_7.pth b/prompt-a/twochar/checkpoint-50/rng_state_7.pth new file mode 100644 index 0000000000000000000000000000000000000000..00845597001c86526d11ae88103a9edea68ce7f5 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/rng_state_7.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c41b6bd8b9fc31601ff6fd8b87bffa1a1b3486376be9e8e9a6f98f399af08a34 +size 21687 diff --git a/prompt-a/twochar/checkpoint-50/scheduler.pt b/prompt-a/twochar/checkpoint-50/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..a66070eb64faba0589cb72d93620faf364bf42c3 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a13b7fd8d980f681c533668d750003260844bd1cef51c39f5f27763fb2d4c82d +size 627 diff --git a/prompt-a/twochar/checkpoint-50/trainer_state.json b/prompt-a/twochar/checkpoint-50/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..2502a1fdfdaa22740926f7c9a77b59cfd7517a60 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/trainer_state.json @@ -0,0 +1,335 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.5, + "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": 1.9562, + "step": 1 + }, + { + "epoch": 0.02, + "learning_rate": 1.3e-05, + "loss": 2.0273, + "step": 2 + }, + { + "epoch": 0.03, + "learning_rate": 1.9499999999999996e-05, + "loss": 2.2516, + "step": 3 + }, + { + "epoch": 0.04, + "learning_rate": 2.6e-05, + "loss": 2.1955, + "step": 4 + }, + { + "epoch": 0.05, + "learning_rate": 3.25e-05, + "loss": 2.1752, + "step": 5 + }, + { + "epoch": 0.06, + "learning_rate": 3.899999999999999e-05, + "loss": 2.2144, + "step": 6 + }, + { + "epoch": 0.07, + "learning_rate": 4.5499999999999995e-05, + "loss": 2.0829, + "step": 7 + }, + { + "epoch": 0.08, + "learning_rate": 5.2e-05, + "loss": 2.0503, + "step": 8 + }, + { + "epoch": 0.09, + "learning_rate": 5.85e-05, + "loss": 2.1558, + "step": 9 + }, + { + "epoch": 0.1, + "learning_rate": 6.5e-05, + "loss": 2.0924, + "step": 10 + }, + { + "epoch": 0.11, + "learning_rate": 6.499555741225524e-05, + "loss": 2.1238, + "step": 11 + }, + { + "epoch": 0.12, + "learning_rate": 6.498223086358009e-05, + "loss": 2.129, + "step": 12 + }, + { + "epoch": 0.13, + "learning_rate": 6.49600239973199e-05, + "loss": 2.0559, + "step": 13 + }, + { + "epoch": 0.14, + "learning_rate": 6.492894288461016e-05, + "loss": 2.0659, + "step": 14 + }, + { + "epoch": 0.15, + "learning_rate": 6.488899602271677e-05, + "loss": 2.1563, + "step": 15 + }, + { + "epoch": 0.16, + "learning_rate": 6.484019433271287e-05, + "loss": 2.1706, + "step": 16 + }, + { + "epoch": 0.17, + "learning_rate": 6.478255115649324e-05, + "loss": 2.0736, + "step": 17 + }, + { + "epoch": 0.18, + "learning_rate": 6.471608225312668e-05, + "loss": 2.1295, + "step": 18 + }, + { + "epoch": 0.19, + "learning_rate": 6.464080579454772e-05, + "loss": 2.0618, + "step": 19 + }, + { + "epoch": 0.2, + "learning_rate": 6.455674236058847e-05, + "loss": 2.12, + "step": 20 + }, + { + "epoch": 0.2, + "eval_loss": 2.1955978870391846, + "eval_runtime": 2.9627, + "eval_samples_per_second": 3.375, + "eval_steps_per_second": 0.675, + "step": 20 + }, + { + "epoch": 0.21, + "learning_rate": 6.446391493335242e-05, + "loss": 2.1576, + "step": 21 + }, + { + "epoch": 0.22, + "learning_rate": 6.43623488909313e-05, + "loss": 2.0446, + "step": 22 + }, + { + "epoch": 0.23, + "learning_rate": 6.425207200046697e-05, + "loss": 2.0033, + "step": 23 + }, + { + "epoch": 0.24, + "learning_rate": 6.413311441056019e-05, + "loss": 2.1382, + "step": 24 + }, + { + "epoch": 0.25, + "learning_rate": 6.400550864302823e-05, + "loss": 2.0989, + "step": 25 + }, + { + "epoch": 0.26, + "learning_rate": 6.386928958401383e-05, + "loss": 2.1276, + "step": 26 + }, + { + "epoch": 0.27, + "learning_rate": 6.372449447444757e-05, + "loss": 2.023, + "step": 27 + }, + { + "epoch": 0.28, + "learning_rate": 6.357116289986662e-05, + "loss": 2.0597, + "step": 28 + }, + { + "epoch": 0.29, + "learning_rate": 6.340933677959248e-05, + "loss": 2.0529, + "step": 29 + }, + { + "epoch": 0.3, + "learning_rate": 6.323906035527062e-05, + "loss": 1.9873, + "step": 30 + }, + { + "epoch": 0.31, + "learning_rate": 6.306038017877523e-05, + "loss": 2.0613, + "step": 31 + }, + { + "epoch": 0.32, + "learning_rate": 6.287334509948247e-05, + "loss": 2.0765, + "step": 32 + }, + { + "epoch": 0.33, + "learning_rate": 6.267800625091547e-05, + "loss": 2.0809, + "step": 33 + }, + { + "epoch": 0.34, + "learning_rate": 6.2474417036765e-05, + "loss": 2.072, + "step": 34 + }, + { + "epoch": 0.35, + "learning_rate": 6.226263311628936e-05, + "loss": 2.0069, + "step": 35 + }, + { + "epoch": 0.36, + "learning_rate": 6.204271238909777e-05, + "loss": 2.0784, + "step": 36 + }, + { + "epoch": 0.37, + "learning_rate": 6.181471497932117e-05, + "loss": 2.1763, + "step": 37 + }, + { + "epoch": 0.38, + "learning_rate": 6.157870321917481e-05, + "loss": 2.0402, + "step": 38 + }, + { + "epoch": 0.39, + "learning_rate": 6.133474163191738e-05, + "loss": 2.1082, + "step": 39 + }, + { + "epoch": 0.4, + "learning_rate": 6.108289691421088e-05, + "loss": 2.1381, + "step": 40 + }, + { + "epoch": 0.4, + "eval_loss": 2.172107219696045, + "eval_runtime": 2.97, + "eval_samples_per_second": 3.367, + "eval_steps_per_second": 0.673, + "step": 40 + }, + { + "epoch": 0.41, + "learning_rate": 6.08232379178865e-05, + "loss": 2.0538, + "step": 41 + }, + { + "epoch": 0.42, + "learning_rate": 6.055583563112115e-05, + "loss": 2.222, + "step": 42 + }, + { + "epoch": 0.43, + "learning_rate": 6.0280763159030015e-05, + "loss": 1.996, + "step": 43 + }, + { + "epoch": 0.44, + "learning_rate": 5.999809570368039e-05, + "loss": 1.9798, + "step": 44 + }, + { + "epoch": 0.45, + "learning_rate": 5.970791054353217e-05, + "loss": 2.1825, + "step": 45 + }, + { + "epoch": 0.46, + "learning_rate": 5.941028701231065e-05, + "loss": 2.1671, + "step": 46 + }, + { + "epoch": 0.47, + "learning_rate": 5.91053064773175e-05, + "loss": 2.0992, + "step": 47 + }, + { + "epoch": 0.48, + "learning_rate": 5.8793052317185785e-05, + "loss": 2.0544, + "step": 48 + }, + { + "epoch": 0.49, + "learning_rate": 5.8473609899085046e-05, + "loss": 2.1054, + "step": 49 + }, + { + "epoch": 0.5, + "learning_rate": 5.814706655538279e-05, + "loss": 2.1214, + "step": 50 + } + ], + "logging_steps": 1, + "max_steps": 200, + "num_train_epochs": 2, + "save_steps": 50, + "total_flos": 4.206166558742938e+16, + "trial_name": null, + "trial_params": null +} diff --git a/prompt-a/twochar/checkpoint-50/training_args.bin b/prompt-a/twochar/checkpoint-50/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..d58072f53f91b75b135f9498c19f7bcc5b637ba4 --- /dev/null +++ b/prompt-a/twochar/checkpoint-50/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9212c1ed1ebf30dd523c0f488bee8e1daf66e67eaaf76d60029260d57f02f88 +size 4411 diff --git a/prompt-a/twochar/config.json b/prompt-a/twochar/config.json new file mode 100644 index 0000000000000000000000000000000000000000..99be7fbf147c96ef93ed720729ea90e8e21bdfca --- /dev/null +++ b/prompt-a/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/prompt-a/twochar/special_tokens_map.json b/prompt-a/twochar/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..3f58a5e115855c6ea3cec98accae196ad927222e --- /dev/null +++ b/prompt-a/twochar/special_tokens_map.json @@ -0,0 +1,6 @@ +{ + "bos_token": "", + "eos_token": "", + "pad_token": "[PAD]", + "unk_token": "" +} diff --git a/prompt-a/twochar/tokenizer.model b/prompt-a/twochar/tokenizer.model new file mode 100644 index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899 --- /dev/null +++ b/prompt-a/twochar/tokenizer.model @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347 +size 499723 diff --git a/prompt-a/twochar/tokenizer_config.json b/prompt-a/twochar/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..77849a9a3ca128f00bc3febdc9b39391a4ef43a3 --- /dev/null +++ b/prompt-a/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 +}