Commit ·
f070e92
1
Parent(s): 12b036a
Upload folder using huggingface_hub
Browse files- README.md +21 -0
- adapter_config.json +21 -0
- adapter_model.bin +3 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer.model +3 -0
- tokenizer_config.json +34 -0
- train_args.json +123 -0
- training_args.bin +3 -0
- training_logs.json +1067 -0
README.md
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- load_in_8bit: False
|
| 10 |
+
- load_in_4bit: True
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: nf4
|
| 16 |
+
- bnb_4bit_use_double_quant: True
|
| 17 |
+
- bnb_4bit_compute_dtype: float16
|
| 18 |
+
### Framework versions
|
| 19 |
+
|
| 20 |
+
|
| 21 |
+
- PEFT 0.5.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "models/openllama-3b",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 16,
|
| 11 |
+
"lora_dropout": 0.1,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 64,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "SEQ_CLS"
|
| 21 |
+
}
|
adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aaffa5b670010b770a585a6062052ba6f0fe7ba9b2aa88948f104809a383a35c
|
| 3 |
+
size 85258545
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": true,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab1b681ec7fc02fed5edd3026687d7a692a918c4dd8e150ca2e3994a6229843b
|
| 3 |
+
size 534194
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"__type": "AddedToken",
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": true,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
"clean_up_tokenization_spaces": false,
|
| 11 |
+
"eos_token": {
|
| 12 |
+
"__type": "AddedToken",
|
| 13 |
+
"content": "</s>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false
|
| 18 |
+
},
|
| 19 |
+
"legacy": null,
|
| 20 |
+
"model_max_length": 2048,
|
| 21 |
+
"pad_token": null,
|
| 22 |
+
"sp_model_kwargs": {},
|
| 23 |
+
"spaces_between_special_tokens": false,
|
| 24 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 25 |
+
"unk_token": {
|
| 26 |
+
"__type": "AddedToken",
|
| 27 |
+
"content": "<unk>",
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"normalized": true,
|
| 30 |
+
"rstrip": false,
|
| 31 |
+
"single_word": false
|
| 32 |
+
},
|
| 33 |
+
"use_default_system_prompt": true
|
| 34 |
+
}
|
train_args.json
ADDED
|
@@ -0,0 +1,123 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"initial_model_dir": "models/openllama-3b",
|
| 3 |
+
"distribution_id": "math",
|
| 4 |
+
"date_trained": "15/10/2023 08:02:52",
|
| 5 |
+
"output_dir": "models/classify_lora/openllama-3b-math",
|
| 6 |
+
"overwrite_output_dir": false,
|
| 7 |
+
"do_train": false,
|
| 8 |
+
"do_eval": true,
|
| 9 |
+
"do_predict": false,
|
| 10 |
+
"evaluation_strategy": "steps",
|
| 11 |
+
"prediction_loss_only": false,
|
| 12 |
+
"per_device_train_batch_size": 16,
|
| 13 |
+
"per_device_eval_batch_size": 32,
|
| 14 |
+
"per_gpu_train_batch_size": null,
|
| 15 |
+
"per_gpu_eval_batch_size": null,
|
| 16 |
+
"gradient_accumulation_steps": 1,
|
| 17 |
+
"eval_accumulation_steps": null,
|
| 18 |
+
"eval_delay": 0,
|
| 19 |
+
"learning_rate": 7.2e-05,
|
| 20 |
+
"weight_decay": 0.0,
|
| 21 |
+
"adam_beta1": 0.9,
|
| 22 |
+
"adam_beta2": 0.999,
|
| 23 |
+
"adam_epsilon": 1e-08,
|
| 24 |
+
"max_grad_norm": 0.3,
|
| 25 |
+
"num_train_epochs": 2,
|
| 26 |
+
"max_steps": 100,
|
| 27 |
+
"lr_scheduler_type": "constant",
|
| 28 |
+
"warmup_ratio": 0.03,
|
| 29 |
+
"warmup_steps": 0,
|
| 30 |
+
"log_level": "passive",
|
| 31 |
+
"log_level_replica": "warning",
|
| 32 |
+
"log_on_each_node": true,
|
| 33 |
+
"logging_dir": "models/classify_lora/openllama-3b-math/runs/Oct15_07-27-16_compute-permanent-node-990",
|
| 34 |
+
"logging_strategy": "steps",
|
| 35 |
+
"logging_first_step": false,
|
| 36 |
+
"logging_steps": 1,
|
| 37 |
+
"logging_nan_inf_filter": true,
|
| 38 |
+
"save_strategy": "steps",
|
| 39 |
+
"save_steps": 25,
|
| 40 |
+
"save_total_limit": 0,
|
| 41 |
+
"save_safetensors": false,
|
| 42 |
+
"save_on_each_node": false,
|
| 43 |
+
"no_cuda": false,
|
| 44 |
+
"use_cpu": false,
|
| 45 |
+
"use_mps_device": false,
|
| 46 |
+
"seed": 42,
|
| 47 |
+
"data_seed": null,
|
| 48 |
+
"jit_mode_eval": false,
|
| 49 |
+
"use_ipex": false,
|
| 50 |
+
"bf16": false,
|
| 51 |
+
"fp16": false,
|
| 52 |
+
"fp16_opt_level": "O1",
|
| 53 |
+
"half_precision_backend": "auto",
|
| 54 |
+
"bf16_full_eval": false,
|
| 55 |
+
"fp16_full_eval": false,
|
| 56 |
+
"tf32": null,
|
| 57 |
+
"local_rank": 0,
|
| 58 |
+
"ddp_backend": null,
|
| 59 |
+
"tpu_num_cores": null,
|
| 60 |
+
"tpu_metrics_debug": false,
|
| 61 |
+
"debug": [],
|
| 62 |
+
"dataloader_drop_last": false,
|
| 63 |
+
"eval_steps": 25,
|
| 64 |
+
"dataloader_num_workers": 0,
|
| 65 |
+
"past_index": -1,
|
| 66 |
+
"run_name": "train|models-classify_lora-openllama-3b-math",
|
| 67 |
+
"disable_tqdm": false,
|
| 68 |
+
"remove_unused_columns": false,
|
| 69 |
+
"label_names": null,
|
| 70 |
+
"load_best_model_at_end": false,
|
| 71 |
+
"metric_for_best_model": "eval_math_score",
|
| 72 |
+
"greater_is_better": true,
|
| 73 |
+
"ignore_data_skip": false,
|
| 74 |
+
"sharded_ddp": [],
|
| 75 |
+
"fsdp": [],
|
| 76 |
+
"fsdp_min_num_params": 0,
|
| 77 |
+
"fsdp_config": {
|
| 78 |
+
"min_num_params": 0,
|
| 79 |
+
"xla": false,
|
| 80 |
+
"xla_fsdp_grad_ckpt": false
|
| 81 |
+
},
|
| 82 |
+
"fsdp_transformer_layer_cls_to_wrap": null,
|
| 83 |
+
"deepspeed": "configs/ds_zero_1.json",
|
| 84 |
+
"label_smoothing_factor": 0.0,
|
| 85 |
+
"optim": "paged_adamw_32bit",
|
| 86 |
+
"optim_args": null,
|
| 87 |
+
"adafactor": false,
|
| 88 |
+
"group_by_length": false,
|
| 89 |
+
"length_column_name": "length",
|
| 90 |
+
"report_to": [
|
| 91 |
+
"wandb"
|
| 92 |
+
],
|
| 93 |
+
"ddp_find_unused_parameters": false,
|
| 94 |
+
"ddp_bucket_cap_mb": null,
|
| 95 |
+
"ddp_broadcast_buffers": null,
|
| 96 |
+
"dataloader_pin_memory": true,
|
| 97 |
+
"skip_memory_metrics": true,
|
| 98 |
+
"use_legacy_prediction_loop": false,
|
| 99 |
+
"push_to_hub": false,
|
| 100 |
+
"resume_from_checkpoint": null,
|
| 101 |
+
"hub_model_id": null,
|
| 102 |
+
"hub_strategy": "every_save",
|
| 103 |
+
"hub_token": null,
|
| 104 |
+
"hub_private_repo": false,
|
| 105 |
+
"hub_always_push": false,
|
| 106 |
+
"gradient_checkpointing": false,
|
| 107 |
+
"include_inputs_for_metrics": false,
|
| 108 |
+
"fp16_backend": "auto",
|
| 109 |
+
"push_to_hub_model_id": null,
|
| 110 |
+
"push_to_hub_organization": null,
|
| 111 |
+
"push_to_hub_token": null,
|
| 112 |
+
"_n_gpu": 1,
|
| 113 |
+
"mp_parameters": "",
|
| 114 |
+
"auto_find_batch_size": false,
|
| 115 |
+
"full_determinism": false,
|
| 116 |
+
"torchdynamo": null,
|
| 117 |
+
"ray_scope": "last",
|
| 118 |
+
"ddp_timeout": 1800,
|
| 119 |
+
"torch_compile": false,
|
| 120 |
+
"torch_compile_backend": null,
|
| 121 |
+
"torch_compile_mode": null,
|
| 122 |
+
"dispatch_batches": null
|
| 123 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:35ec9fe354e1c52684da09de61aeb8d1787eb721240a5ef8b1035e4f3ac721bd
|
| 3 |
+
size 5115
|
training_logs.json
ADDED
|
@@ -0,0 +1,1067 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"loss": 0.6862,
|
| 4 |
+
"learning_rate": 7.2e-05,
|
| 5 |
+
"epoch": 0.02,
|
| 6 |
+
"step": 1
|
| 7 |
+
},
|
| 8 |
+
{
|
| 9 |
+
"loss": 0.692,
|
| 10 |
+
"learning_rate": 7.2e-05,
|
| 11 |
+
"epoch": 0.04,
|
| 12 |
+
"step": 2
|
| 13 |
+
},
|
| 14 |
+
{
|
| 15 |
+
"loss": 0.6954,
|
| 16 |
+
"learning_rate": 7.2e-05,
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"step": 3
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"loss": 0.7031,
|
| 22 |
+
"learning_rate": 7.2e-05,
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
+
"step": 4
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"loss": 0.6928,
|
| 28 |
+
"learning_rate": 7.2e-05,
|
| 29 |
+
"epoch": 0.09,
|
| 30 |
+
"step": 5
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"loss": 0.699,
|
| 34 |
+
"learning_rate": 7.2e-05,
|
| 35 |
+
"epoch": 0.11,
|
| 36 |
+
"step": 6
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"loss": 0.6902,
|
| 40 |
+
"learning_rate": 7.2e-05,
|
| 41 |
+
"epoch": 0.12,
|
| 42 |
+
"step": 7
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"loss": 0.681,
|
| 46 |
+
"learning_rate": 7.2e-05,
|
| 47 |
+
"epoch": 0.14,
|
| 48 |
+
"step": 8
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"loss": 0.6931,
|
| 52 |
+
"learning_rate": 7.2e-05,
|
| 53 |
+
"epoch": 0.16,
|
| 54 |
+
"step": 9
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"loss": 0.6892,
|
| 58 |
+
"learning_rate": 7.2e-05,
|
| 59 |
+
"epoch": 0.18,
|
| 60 |
+
"step": 10
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"loss": 0.6898,
|
| 64 |
+
"learning_rate": 7.2e-05,
|
| 65 |
+
"epoch": 0.19,
|
| 66 |
+
"step": 11
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"loss": 0.6982,
|
| 70 |
+
"learning_rate": 7.2e-05,
|
| 71 |
+
"epoch": 0.21,
|
| 72 |
+
"step": 12
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"loss": 0.6922,
|
| 76 |
+
"learning_rate": 7.2e-05,
|
| 77 |
+
"epoch": 0.23,
|
| 78 |
+
"step": 13
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"loss": 0.6865,
|
| 82 |
+
"learning_rate": 7.2e-05,
|
| 83 |
+
"epoch": 0.25,
|
| 84 |
+
"step": 14
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"loss": 0.6901,
|
| 88 |
+
"learning_rate": 7.2e-05,
|
| 89 |
+
"epoch": 0.26,
|
| 90 |
+
"step": 15
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"loss": 0.6953,
|
| 94 |
+
"learning_rate": 7.2e-05,
|
| 95 |
+
"epoch": 0.28,
|
| 96 |
+
"step": 16
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"loss": 0.6845,
|
| 100 |
+
"learning_rate": 7.2e-05,
|
| 101 |
+
"epoch": 0.3,
|
| 102 |
+
"step": 17
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"loss": 0.6914,
|
| 106 |
+
"learning_rate": 7.2e-05,
|
| 107 |
+
"epoch": 0.32,
|
| 108 |
+
"step": 18
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"loss": 0.6885,
|
| 112 |
+
"learning_rate": 7.2e-05,
|
| 113 |
+
"epoch": 0.33,
|
| 114 |
+
"step": 19
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"loss": 0.6903,
|
| 118 |
+
"learning_rate": 7.2e-05,
|
| 119 |
+
"epoch": 0.35,
|
| 120 |
+
"step": 20
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"loss": 0.6895,
|
| 124 |
+
"learning_rate": 7.2e-05,
|
| 125 |
+
"epoch": 0.37,
|
| 126 |
+
"step": 21
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"loss": 0.6775,
|
| 130 |
+
"learning_rate": 7.2e-05,
|
| 131 |
+
"epoch": 0.39,
|
| 132 |
+
"step": 22
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"loss": 0.6843,
|
| 136 |
+
"learning_rate": 7.2e-05,
|
| 137 |
+
"epoch": 0.4,
|
| 138 |
+
"step": 23
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"loss": 0.6778,
|
| 142 |
+
"learning_rate": 7.2e-05,
|
| 143 |
+
"epoch": 0.42,
|
| 144 |
+
"step": 24
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"loss": 0.6954,
|
| 148 |
+
"learning_rate": 7.2e-05,
|
| 149 |
+
"epoch": 0.44,
|
| 150 |
+
"step": 25
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"eval_math_loss": 0.6901352405548096,
|
| 154 |
+
"eval_math_score": -0.25167757272720337,
|
| 155 |
+
"eval_math_brier_score": 0.25167757272720337,
|
| 156 |
+
"eval_math_average_probability": 0.4987456500530243,
|
| 157 |
+
"eval_math_accuracy": 0.62,
|
| 158 |
+
"eval_math_probabilities": [
|
| 159 |
+
0.5027638673782349,
|
| 160 |
+
0.5069999694824219,
|
| 161 |
+
0.4998233914375305,
|
| 162 |
+
0.49942639470100403,
|
| 163 |
+
0.5001113414764404,
|
| 164 |
+
0.4992215633392334,
|
| 165 |
+
0.5041193962097168,
|
| 166 |
+
0.49987202882766724,
|
| 167 |
+
0.5148451328277588,
|
| 168 |
+
0.525566041469574,
|
| 169 |
+
0.5015634894371033,
|
| 170 |
+
0.5299053192138672,
|
| 171 |
+
0.500709056854248,
|
| 172 |
+
0.49956902861595154,
|
| 173 |
+
0.5017619729042053,
|
| 174 |
+
0.5067934989929199,
|
| 175 |
+
0.4978564977645874,
|
| 176 |
+
0.4987451434135437,
|
| 177 |
+
0.4950011372566223,
|
| 178 |
+
0.5006818175315857,
|
| 179 |
+
0.4997928738594055,
|
| 180 |
+
0.501230776309967,
|
| 181 |
+
0.5018536448478699,
|
| 182 |
+
0.5026345252990723,
|
| 183 |
+
0.4967585802078247,
|
| 184 |
+
0.4941774606704712,
|
| 185 |
+
0.4975478947162628,
|
| 186 |
+
0.5060742497444153,
|
| 187 |
+
0.47517138719558716,
|
| 188 |
+
0.5136992931365967,
|
| 189 |
+
0.5003398060798645,
|
| 190 |
+
0.500635027885437,
|
| 191 |
+
0.5026965737342834,
|
| 192 |
+
0.49716416001319885,
|
| 193 |
+
0.5036399960517883,
|
| 194 |
+
0.5027968287467957,
|
| 195 |
+
0.4978507161140442,
|
| 196 |
+
0.49902665615081787,
|
| 197 |
+
0.49480193853378296,
|
| 198 |
+
0.5012259483337402,
|
| 199 |
+
0.4997442662715912,
|
| 200 |
+
0.49923667311668396,
|
| 201 |
+
0.4908088147640228,
|
| 202 |
+
0.5081710815429688,
|
| 203 |
+
0.4765812158584595,
|
| 204 |
+
0.5016435384750366,
|
| 205 |
+
0.5013003945350647,
|
| 206 |
+
0.4970977306365967,
|
| 207 |
+
0.5095471143722534,
|
| 208 |
+
0.4940992593765259,
|
| 209 |
+
0.5008733868598938,
|
| 210 |
+
0.5002974271774292,
|
| 211 |
+
0.5011077523231506,
|
| 212 |
+
0.49904587864875793,
|
| 213 |
+
0.5007302761077881,
|
| 214 |
+
0.5004144310951233,
|
| 215 |
+
0.5012476444244385,
|
| 216 |
+
0.5047202706336975,
|
| 217 |
+
0.5079423785209656,
|
| 218 |
+
0.5031242370605469,
|
| 219 |
+
0.5022978186607361,
|
| 220 |
+
0.4994467496871948,
|
| 221 |
+
0.5034329891204834,
|
| 222 |
+
0.5073589086532593,
|
| 223 |
+
0.5306907892227173,
|
| 224 |
+
0.5277766585350037,
|
| 225 |
+
0.500605583190918,
|
| 226 |
+
0.49940890073776245,
|
| 227 |
+
0.5001015067100525,
|
| 228 |
+
0.5020729899406433,
|
| 229 |
+
0.49872809648513794,
|
| 230 |
+
0.5019404888153076,
|
| 231 |
+
0.5015720129013062,
|
| 232 |
+
0.505891740322113,
|
| 233 |
+
0.5065779685974121,
|
| 234 |
+
0.5094985961914062,
|
| 235 |
+
0.48333850502967834,
|
| 236 |
+
0.4985012710094452,
|
| 237 |
+
0.5102980732917786,
|
| 238 |
+
0.5111819505691528,
|
| 239 |
+
0.5107572078704834,
|
| 240 |
+
0.5016093850135803,
|
| 241 |
+
0.5002478361129761,
|
| 242 |
+
0.5022361874580383,
|
| 243 |
+
0.5004898309707642,
|
| 244 |
+
0.4977990686893463,
|
| 245 |
+
0.49885597825050354,
|
| 246 |
+
0.5022542476654053,
|
| 247 |
+
0.49431321024894714,
|
| 248 |
+
0.5005004405975342,
|
| 249 |
+
0.5246015787124634,
|
| 250 |
+
0.5063254833221436,
|
| 251 |
+
0.505049467086792,
|
| 252 |
+
0.4940093457698822,
|
| 253 |
+
0.49063223600387573,
|
| 254 |
+
0.4952346384525299,
|
| 255 |
+
0.3910456895828247,
|
| 256 |
+
0.39186742901802063,
|
| 257 |
+
0.3929601311683655,
|
| 258 |
+
0.5008682012557983
|
| 259 |
+
],
|
| 260 |
+
"eval_math_runtime": 22.613,
|
| 261 |
+
"eval_math_samples_per_second": 4.422,
|
| 262 |
+
"eval_math_steps_per_second": 0.088,
|
| 263 |
+
"epoch": 0.44,
|
| 264 |
+
"step": 25
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"loss": 0.6919,
|
| 268 |
+
"learning_rate": 7.2e-05,
|
| 269 |
+
"epoch": 0.46,
|
| 270 |
+
"step": 26
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"loss": 0.6883,
|
| 274 |
+
"learning_rate": 7.2e-05,
|
| 275 |
+
"epoch": 0.47,
|
| 276 |
+
"step": 27
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"loss": 0.6733,
|
| 280 |
+
"learning_rate": 7.2e-05,
|
| 281 |
+
"epoch": 0.49,
|
| 282 |
+
"step": 28
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"loss": 0.6873,
|
| 286 |
+
"learning_rate": 7.2e-05,
|
| 287 |
+
"epoch": 0.51,
|
| 288 |
+
"step": 29
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"loss": 0.6645,
|
| 292 |
+
"learning_rate": 7.2e-05,
|
| 293 |
+
"epoch": 0.53,
|
| 294 |
+
"step": 30
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"loss": 0.6799,
|
| 298 |
+
"learning_rate": 7.2e-05,
|
| 299 |
+
"epoch": 0.54,
|
| 300 |
+
"step": 31
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"loss": 0.6721,
|
| 304 |
+
"learning_rate": 7.2e-05,
|
| 305 |
+
"epoch": 0.56,
|
| 306 |
+
"step": 32
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"loss": 0.6906,
|
| 310 |
+
"learning_rate": 7.2e-05,
|
| 311 |
+
"epoch": 0.58,
|
| 312 |
+
"step": 33
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"loss": 0.6775,
|
| 316 |
+
"learning_rate": 7.2e-05,
|
| 317 |
+
"epoch": 0.6,
|
| 318 |
+
"step": 34
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"loss": 0.6868,
|
| 322 |
+
"learning_rate": 7.2e-05,
|
| 323 |
+
"epoch": 0.61,
|
| 324 |
+
"step": 35
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"loss": 0.6691,
|
| 328 |
+
"learning_rate": 7.2e-05,
|
| 329 |
+
"epoch": 0.63,
|
| 330 |
+
"step": 36
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"loss": 0.6775,
|
| 334 |
+
"learning_rate": 7.2e-05,
|
| 335 |
+
"epoch": 0.65,
|
| 336 |
+
"step": 37
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"loss": 0.6875,
|
| 340 |
+
"learning_rate": 7.2e-05,
|
| 341 |
+
"epoch": 0.67,
|
| 342 |
+
"step": 38
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"loss": 0.6882,
|
| 346 |
+
"learning_rate": 7.2e-05,
|
| 347 |
+
"epoch": 0.68,
|
| 348 |
+
"step": 39
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"loss": 0.6659,
|
| 352 |
+
"learning_rate": 7.2e-05,
|
| 353 |
+
"epoch": 0.7,
|
| 354 |
+
"step": 40
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"loss": 0.6831,
|
| 358 |
+
"learning_rate": 7.2e-05,
|
| 359 |
+
"epoch": 0.72,
|
| 360 |
+
"step": 41
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"loss": 0.6883,
|
| 364 |
+
"learning_rate": 7.2e-05,
|
| 365 |
+
"epoch": 0.74,
|
| 366 |
+
"step": 42
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"loss": 0.6687,
|
| 370 |
+
"learning_rate": 7.2e-05,
|
| 371 |
+
"epoch": 0.75,
|
| 372 |
+
"step": 43
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"loss": 0.6756,
|
| 376 |
+
"learning_rate": 7.2e-05,
|
| 377 |
+
"epoch": 0.77,
|
| 378 |
+
"step": 44
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"loss": 0.6767,
|
| 382 |
+
"learning_rate": 7.2e-05,
|
| 383 |
+
"epoch": 0.79,
|
| 384 |
+
"step": 45
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"loss": 0.6726,
|
| 388 |
+
"learning_rate": 7.2e-05,
|
| 389 |
+
"epoch": 0.81,
|
| 390 |
+
"step": 46
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"loss": 0.6817,
|
| 394 |
+
"learning_rate": 7.2e-05,
|
| 395 |
+
"epoch": 0.82,
|
| 396 |
+
"step": 47
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"loss": 0.6843,
|
| 400 |
+
"learning_rate": 7.2e-05,
|
| 401 |
+
"epoch": 0.84,
|
| 402 |
+
"step": 48
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"loss": 0.6677,
|
| 406 |
+
"learning_rate": 7.2e-05,
|
| 407 |
+
"epoch": 0.86,
|
| 408 |
+
"step": 49
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"loss": 0.675,
|
| 412 |
+
"learning_rate": 7.2e-05,
|
| 413 |
+
"epoch": 0.88,
|
| 414 |
+
"step": 50
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"eval_math_loss": 0.6834437847137451,
|
| 418 |
+
"eval_math_score": -0.25087547302246094,
|
| 419 |
+
"eval_math_brier_score": 0.25087547302246094,
|
| 420 |
+
"eval_math_average_probability": 0.5003729462623596,
|
| 421 |
+
"eval_math_accuracy": 0.67,
|
| 422 |
+
"eval_math_probabilities": [
|
| 423 |
+
0.5053072571754456,
|
| 424 |
+
0.5104738473892212,
|
| 425 |
+
0.4998767375946045,
|
| 426 |
+
0.4987499713897705,
|
| 427 |
+
0.4995637536048889,
|
| 428 |
+
0.4979211091995239,
|
| 429 |
+
0.5108855366706848,
|
| 430 |
+
0.5023437142372131,
|
| 431 |
+
0.5171610713005066,
|
| 432 |
+
0.5161605477333069,
|
| 433 |
+
0.5233566761016846,
|
| 434 |
+
0.5833942294120789,
|
| 435 |
+
0.5003718733787537,
|
| 436 |
+
0.49910590052604675,
|
| 437 |
+
0.5014094114303589,
|
| 438 |
+
0.5064855217933655,
|
| 439 |
+
0.49743178486824036,
|
| 440 |
+
0.4980136752128601,
|
| 441 |
+
0.4959304630756378,
|
| 442 |
+
0.5030016899108887,
|
| 443 |
+
0.5019874572753906,
|
| 444 |
+
0.5007861852645874,
|
| 445 |
+
0.5035613179206848,
|
| 446 |
+
0.5047624707221985,
|
| 447 |
+
0.49380984902381897,
|
| 448 |
+
0.4902263581752777,
|
| 449 |
+
0.49334290623664856,
|
| 450 |
+
0.5064259767532349,
|
| 451 |
+
0.4730111360549927,
|
| 452 |
+
0.5091708302497864,
|
| 453 |
+
0.5004432201385498,
|
| 454 |
+
0.49959176778793335,
|
| 455 |
+
0.4988599419593811,
|
| 456 |
+
0.5252774357795715,
|
| 457 |
+
0.5397801399230957,
|
| 458 |
+
0.5378824472427368,
|
| 459 |
+
0.5073467493057251,
|
| 460 |
+
0.5127724409103394,
|
| 461 |
+
0.48429790139198303,
|
| 462 |
+
0.5011600852012634,
|
| 463 |
+
0.499787300825119,
|
| 464 |
+
0.5010166168212891,
|
| 465 |
+
0.49625164270401,
|
| 466 |
+
0.512554407119751,
|
| 467 |
+
0.48166966438293457,
|
| 468 |
+
0.5029576420783997,
|
| 469 |
+
0.49886441230773926,
|
| 470 |
+
0.49318164587020874,
|
| 471 |
+
0.5096920728683472,
|
| 472 |
+
0.49129679799079895,
|
| 473 |
+
0.5006589889526367,
|
| 474 |
+
0.5004881024360657,
|
| 475 |
+
0.5014761090278625,
|
| 476 |
+
0.500096321105957,
|
| 477 |
+
0.5061331987380981,
|
| 478 |
+
0.5045473575592041,
|
| 479 |
+
0.506636381149292,
|
| 480 |
+
0.510319173336029,
|
| 481 |
+
0.5139249563217163,
|
| 482 |
+
0.516427218914032,
|
| 483 |
+
0.5025309324264526,
|
| 484 |
+
0.5002322196960449,
|
| 485 |
+
0.49999839067459106,
|
| 486 |
+
0.5077803730964661,
|
| 487 |
+
0.5266351103782654,
|
| 488 |
+
0.5349680185317993,
|
| 489 |
+
0.5012303590774536,
|
| 490 |
+
0.5011308789253235,
|
| 491 |
+
0.502089262008667,
|
| 492 |
+
0.49999287724494934,
|
| 493 |
+
0.49622201919555664,
|
| 494 |
+
0.5034714937210083,
|
| 495 |
+
0.5027754306793213,
|
| 496 |
+
0.5089088678359985,
|
| 497 |
+
0.5117012858390808,
|
| 498 |
+
0.5095623135566711,
|
| 499 |
+
0.4832465946674347,
|
| 500 |
+
0.49723362922668457,
|
| 501 |
+
0.5502822399139404,
|
| 502 |
+
0.5483925342559814,
|
| 503 |
+
0.5437333583831787,
|
| 504 |
+
0.5012779831886292,
|
| 505 |
+
0.4989190697669983,
|
| 506 |
+
0.5021101236343384,
|
| 507 |
+
0.5062341094017029,
|
| 508 |
+
0.5053580403327942,
|
| 509 |
+
0.5082789659500122,
|
| 510 |
+
0.5026023387908936,
|
| 511 |
+
0.49390891194343567,
|
| 512 |
+
0.5006174445152283,
|
| 513 |
+
0.5223895311355591,
|
| 514 |
+
0.5041326284408569,
|
| 515 |
+
0.5024306178092957,
|
| 516 |
+
0.492806077003479,
|
| 517 |
+
0.513796865940094,
|
| 518 |
+
0.4959098994731903,
|
| 519 |
+
0.31710273027420044,
|
| 520 |
+
0.3174270689487457,
|
| 521 |
+
0.3193003535270691,
|
| 522 |
+
0.5011496543884277
|
| 523 |
+
],
|
| 524 |
+
"eval_math_runtime": 22.5936,
|
| 525 |
+
"eval_math_samples_per_second": 4.426,
|
| 526 |
+
"eval_math_steps_per_second": 0.089,
|
| 527 |
+
"epoch": 0.88,
|
| 528 |
+
"step": 50
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"loss": 0.6771,
|
| 532 |
+
"learning_rate": 7.2e-05,
|
| 533 |
+
"epoch": 0.89,
|
| 534 |
+
"step": 51
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"loss": 0.6825,
|
| 538 |
+
"learning_rate": 7.2e-05,
|
| 539 |
+
"epoch": 0.91,
|
| 540 |
+
"step": 52
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"loss": 0.6914,
|
| 544 |
+
"learning_rate": 7.2e-05,
|
| 545 |
+
"epoch": 0.93,
|
| 546 |
+
"step": 53
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"loss": 0.6884,
|
| 550 |
+
"learning_rate": 7.2e-05,
|
| 551 |
+
"epoch": 0.95,
|
| 552 |
+
"step": 54
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"loss": 0.6772,
|
| 556 |
+
"learning_rate": 7.2e-05,
|
| 557 |
+
"epoch": 0.96,
|
| 558 |
+
"step": 55
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"loss": 0.6635,
|
| 562 |
+
"learning_rate": 7.2e-05,
|
| 563 |
+
"epoch": 0.98,
|
| 564 |
+
"step": 56
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"loss": 0.6667,
|
| 568 |
+
"learning_rate": 7.2e-05,
|
| 569 |
+
"epoch": 1.0,
|
| 570 |
+
"step": 57
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"loss": 0.6587,
|
| 574 |
+
"learning_rate": 7.2e-05,
|
| 575 |
+
"epoch": 1.02,
|
| 576 |
+
"step": 58
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"loss": 0.6802,
|
| 580 |
+
"learning_rate": 7.2e-05,
|
| 581 |
+
"epoch": 1.04,
|
| 582 |
+
"step": 59
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"loss": 0.6705,
|
| 586 |
+
"learning_rate": 7.2e-05,
|
| 587 |
+
"epoch": 1.05,
|
| 588 |
+
"step": 60
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"loss": 0.659,
|
| 592 |
+
"learning_rate": 7.2e-05,
|
| 593 |
+
"epoch": 1.07,
|
| 594 |
+
"step": 61
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"loss": 0.6922,
|
| 598 |
+
"learning_rate": 7.2e-05,
|
| 599 |
+
"epoch": 1.09,
|
| 600 |
+
"step": 62
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"loss": 0.6712,
|
| 604 |
+
"learning_rate": 7.2e-05,
|
| 605 |
+
"epoch": 1.11,
|
| 606 |
+
"step": 63
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"loss": 0.6512,
|
| 610 |
+
"learning_rate": 7.2e-05,
|
| 611 |
+
"epoch": 1.12,
|
| 612 |
+
"step": 64
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"loss": 0.6716,
|
| 616 |
+
"learning_rate": 7.2e-05,
|
| 617 |
+
"epoch": 1.14,
|
| 618 |
+
"step": 65
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"loss": 0.6885,
|
| 622 |
+
"learning_rate": 7.2e-05,
|
| 623 |
+
"epoch": 1.16,
|
| 624 |
+
"step": 66
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"loss": 0.6468,
|
| 628 |
+
"learning_rate": 7.2e-05,
|
| 629 |
+
"epoch": 1.18,
|
| 630 |
+
"step": 67
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"loss": 0.6314,
|
| 634 |
+
"learning_rate": 7.2e-05,
|
| 635 |
+
"epoch": 1.19,
|
| 636 |
+
"step": 68
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"loss": 0.6651,
|
| 640 |
+
"learning_rate": 7.2e-05,
|
| 641 |
+
"epoch": 1.21,
|
| 642 |
+
"step": 69
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"loss": 0.6706,
|
| 646 |
+
"learning_rate": 7.2e-05,
|
| 647 |
+
"epoch": 1.23,
|
| 648 |
+
"step": 70
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"loss": 0.6134,
|
| 652 |
+
"learning_rate": 7.2e-05,
|
| 653 |
+
"epoch": 1.25,
|
| 654 |
+
"step": 71
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"loss": 0.6405,
|
| 658 |
+
"learning_rate": 7.2e-05,
|
| 659 |
+
"epoch": 1.26,
|
| 660 |
+
"step": 72
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"loss": 0.6676,
|
| 664 |
+
"learning_rate": 7.2e-05,
|
| 665 |
+
"epoch": 1.28,
|
| 666 |
+
"step": 73
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"loss": 0.6093,
|
| 670 |
+
"learning_rate": 7.2e-05,
|
| 671 |
+
"epoch": 1.3,
|
| 672 |
+
"step": 74
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"loss": 0.6183,
|
| 676 |
+
"learning_rate": 7.2e-05,
|
| 677 |
+
"epoch": 1.32,
|
| 678 |
+
"step": 75
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"eval_math_loss": 0.665087103843689,
|
| 682 |
+
"eval_math_score": -0.24755878746509552,
|
| 683 |
+
"eval_math_brier_score": 0.24755878746509552,
|
| 684 |
+
"eval_math_average_probability": 0.508122980594635,
|
| 685 |
+
"eval_math_accuracy": 0.63,
|
| 686 |
+
"eval_math_probabilities": [
|
| 687 |
+
0.5121186971664429,
|
| 688 |
+
0.5127348899841309,
|
| 689 |
+
0.5004041790962219,
|
| 690 |
+
0.5009912252426147,
|
| 691 |
+
0.5002092123031616,
|
| 692 |
+
0.5010939836502075,
|
| 693 |
+
0.545912504196167,
|
| 694 |
+
0.5255416035652161,
|
| 695 |
+
0.5276244878768921,
|
| 696 |
+
0.5262688994407654,
|
| 697 |
+
0.6037449240684509,
|
| 698 |
+
0.6872636079788208,
|
| 699 |
+
0.45226553082466125,
|
| 700 |
+
0.4923723638057709,
|
| 701 |
+
0.49667802453041077,
|
| 702 |
+
0.5102534890174866,
|
| 703 |
+
0.4977477788925171,
|
| 704 |
+
0.4969833195209503,
|
| 705 |
+
0.49458229541778564,
|
| 706 |
+
0.508873462677002,
|
| 707 |
+
0.5096623301506042,
|
| 708 |
+
0.4953479766845703,
|
| 709 |
+
0.5070879459381104,
|
| 710 |
+
0.5050402283668518,
|
| 711 |
+
0.4883251190185547,
|
| 712 |
+
0.4748927354812622,
|
| 713 |
+
0.47420623898506165,
|
| 714 |
+
0.5024446845054626,
|
| 715 |
+
0.44197729229927063,
|
| 716 |
+
0.5237727165222168,
|
| 717 |
+
0.49600568413734436,
|
| 718 |
+
0.49905824661254883,
|
| 719 |
+
0.4909888207912445,
|
| 720 |
+
0.611699104309082,
|
| 721 |
+
0.656679630279541,
|
| 722 |
+
0.6554825305938721,
|
| 723 |
+
0.5338494777679443,
|
| 724 |
+
0.5640528202056885,
|
| 725 |
+
0.46428170800209045,
|
| 726 |
+
0.4977650046348572,
|
| 727 |
+
0.497366726398468,
|
| 728 |
+
0.508729100227356,
|
| 729 |
+
0.5139797925949097,
|
| 730 |
+
0.5323095321655273,
|
| 731 |
+
0.5103693604469299,
|
| 732 |
+
0.5118908882141113,
|
| 733 |
+
0.493362158536911,
|
| 734 |
+
0.4925827085971832,
|
| 735 |
+
0.5059145092964172,
|
| 736 |
+
0.48735955357551575,
|
| 737 |
+
0.49885982275009155,
|
| 738 |
+
0.5037399530410767,
|
| 739 |
+
0.5033603310585022,
|
| 740 |
+
0.5000958442687988,
|
| 741 |
+
0.5084394216537476,
|
| 742 |
+
0.5043569207191467,
|
| 743 |
+
0.5100818276405334,
|
| 744 |
+
0.5124533772468567,
|
| 745 |
+
0.5260497331619263,
|
| 746 |
+
0.5502150058746338,
|
| 747 |
+
0.5108760595321655,
|
| 748 |
+
0.5024101138114929,
|
| 749 |
+
0.4991067051887512,
|
| 750 |
+
0.49586227536201477,
|
| 751 |
+
0.5228884816169739,
|
| 752 |
+
0.5393039584159851,
|
| 753 |
+
0.49779072403907776,
|
| 754 |
+
0.5032389760017395,
|
| 755 |
+
0.5024223327636719,
|
| 756 |
+
0.5007798671722412,
|
| 757 |
+
0.4911024272441864,
|
| 758 |
+
0.5055527687072754,
|
| 759 |
+
0.5001845359802246,
|
| 760 |
+
0.5026863813400269,
|
| 761 |
+
0.5191549062728882,
|
| 762 |
+
0.5366024971008301,
|
| 763 |
+
0.4879572093486786,
|
| 764 |
+
0.48914453387260437,
|
| 765 |
+
0.6904913187026978,
|
| 766 |
+
0.6880560517311096,
|
| 767 |
+
0.6807676553726196,
|
| 768 |
+
0.4983310401439667,
|
| 769 |
+
0.49475836753845215,
|
| 770 |
+
0.5018371343612671,
|
| 771 |
+
0.5274589657783508,
|
| 772 |
+
0.5387904047966003,
|
| 773 |
+
0.5446135401725769,
|
| 774 |
+
0.49650830030441284,
|
| 775 |
+
0.488943487405777,
|
| 776 |
+
0.4967746436595917,
|
| 777 |
+
0.49659326672554016,
|
| 778 |
+
0.5072319507598877,
|
| 779 |
+
0.505716860294342,
|
| 780 |
+
0.49774613976478577,
|
| 781 |
+
0.553057849407196,
|
| 782 |
+
0.5152707695960999,
|
| 783 |
+
0.17283889651298523,
|
| 784 |
+
0.17193157970905304,
|
| 785 |
+
0.17517608404159546,
|
| 786 |
+
0.5045385956764221
|
| 787 |
+
],
|
| 788 |
+
"eval_math_runtime": 22.5741,
|
| 789 |
+
"eval_math_samples_per_second": 4.43,
|
| 790 |
+
"eval_math_steps_per_second": 0.089,
|
| 791 |
+
"epoch": 1.32,
|
| 792 |
+
"step": 75
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"loss": 0.6424,
|
| 796 |
+
"learning_rate": 7.2e-05,
|
| 797 |
+
"epoch": 1.33,
|
| 798 |
+
"step": 76
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"loss": 0.6144,
|
| 802 |
+
"learning_rate": 7.2e-05,
|
| 803 |
+
"epoch": 1.35,
|
| 804 |
+
"step": 77
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"loss": 0.6389,
|
| 808 |
+
"learning_rate": 7.2e-05,
|
| 809 |
+
"epoch": 1.37,
|
| 810 |
+
"step": 78
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"loss": 0.6363,
|
| 814 |
+
"learning_rate": 7.2e-05,
|
| 815 |
+
"epoch": 1.39,
|
| 816 |
+
"step": 79
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"loss": 0.6334,
|
| 820 |
+
"learning_rate": 7.2e-05,
|
| 821 |
+
"epoch": 1.4,
|
| 822 |
+
"step": 80
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"loss": 0.6685,
|
| 826 |
+
"learning_rate": 7.2e-05,
|
| 827 |
+
"epoch": 1.42,
|
| 828 |
+
"step": 81
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"loss": 0.677,
|
| 832 |
+
"learning_rate": 7.2e-05,
|
| 833 |
+
"epoch": 1.44,
|
| 834 |
+
"step": 82
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"loss": 0.6031,
|
| 838 |
+
"learning_rate": 7.2e-05,
|
| 839 |
+
"epoch": 1.46,
|
| 840 |
+
"step": 83
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"loss": 0.6428,
|
| 844 |
+
"learning_rate": 7.2e-05,
|
| 845 |
+
"epoch": 1.47,
|
| 846 |
+
"step": 84
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"loss": 0.6308,
|
| 850 |
+
"learning_rate": 7.2e-05,
|
| 851 |
+
"epoch": 1.49,
|
| 852 |
+
"step": 85
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"loss": 0.5668,
|
| 856 |
+
"learning_rate": 7.2e-05,
|
| 857 |
+
"epoch": 1.51,
|
| 858 |
+
"step": 86
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"loss": 0.586,
|
| 862 |
+
"learning_rate": 7.2e-05,
|
| 863 |
+
"epoch": 1.53,
|
| 864 |
+
"step": 87
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"loss": 0.6652,
|
| 868 |
+
"learning_rate": 7.2e-05,
|
| 869 |
+
"epoch": 1.54,
|
| 870 |
+
"step": 88
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"loss": 0.5834,
|
| 874 |
+
"learning_rate": 7.2e-05,
|
| 875 |
+
"epoch": 1.56,
|
| 876 |
+
"step": 89
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"loss": 0.6447,
|
| 880 |
+
"learning_rate": 7.2e-05,
|
| 881 |
+
"epoch": 1.58,
|
| 882 |
+
"step": 90
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"loss": 0.5041,
|
| 886 |
+
"learning_rate": 7.2e-05,
|
| 887 |
+
"epoch": 1.6,
|
| 888 |
+
"step": 91
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"loss": 0.5564,
|
| 892 |
+
"learning_rate": 7.2e-05,
|
| 893 |
+
"epoch": 1.61,
|
| 894 |
+
"step": 92
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"loss": 0.5746,
|
| 898 |
+
"learning_rate": 7.2e-05,
|
| 899 |
+
"epoch": 1.63,
|
| 900 |
+
"step": 93
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"loss": 0.6814,
|
| 904 |
+
"learning_rate": 7.2e-05,
|
| 905 |
+
"epoch": 1.65,
|
| 906 |
+
"step": 94
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"loss": 0.6013,
|
| 910 |
+
"learning_rate": 7.2e-05,
|
| 911 |
+
"epoch": 1.67,
|
| 912 |
+
"step": 95
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"loss": 0.5654,
|
| 916 |
+
"learning_rate": 7.2e-05,
|
| 917 |
+
"epoch": 1.68,
|
| 918 |
+
"step": 96
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"loss": 0.5918,
|
| 922 |
+
"learning_rate": 7.2e-05,
|
| 923 |
+
"epoch": 1.7,
|
| 924 |
+
"step": 97
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"loss": 0.5496,
|
| 928 |
+
"learning_rate": 7.2e-05,
|
| 929 |
+
"epoch": 1.72,
|
| 930 |
+
"step": 98
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"loss": 0.6326,
|
| 934 |
+
"learning_rate": 7.2e-05,
|
| 935 |
+
"epoch": 1.74,
|
| 936 |
+
"step": 99
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"loss": 0.5314,
|
| 940 |
+
"learning_rate": 7.2e-05,
|
| 941 |
+
"epoch": 1.75,
|
| 942 |
+
"step": 100
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"eval_math_loss": 0.6169581413269043,
|
| 946 |
+
"eval_math_score": -0.2320980429649353,
|
| 947 |
+
"eval_math_brier_score": 0.2320980429649353,
|
| 948 |
+
"eval_math_average_probability": 0.5478038787841797,
|
| 949 |
+
"eval_math_accuracy": 0.63,
|
| 950 |
+
"eval_math_probabilities": [
|
| 951 |
+
0.5236212611198425,
|
| 952 |
+
0.5186634659767151,
|
| 953 |
+
0.5260717868804932,
|
| 954 |
+
0.489012211561203,
|
| 955 |
+
0.49739015102386475,
|
| 956 |
+
0.4983677864074707,
|
| 957 |
+
0.7061103582382202,
|
| 958 |
+
0.6276401281356812,
|
| 959 |
+
0.5337700843811035,
|
| 960 |
+
0.6074290871620178,
|
| 961 |
+
0.6426130533218384,
|
| 962 |
+
0.8374870419502258,
|
| 963 |
+
0.37604060769081116,
|
| 964 |
+
0.36783960461616516,
|
| 965 |
+
0.39347073435783386,
|
| 966 |
+
0.5029640793800354,
|
| 967 |
+
0.49337878823280334,
|
| 968 |
+
0.4871605336666107,
|
| 969 |
+
0.526091456413269,
|
| 970 |
+
0.510998010635376,
|
| 971 |
+
0.5022907257080078,
|
| 972 |
+
0.5005046129226685,
|
| 973 |
+
0.5276081562042236,
|
| 974 |
+
0.5287099480628967,
|
| 975 |
+
0.5003510117530823,
|
| 976 |
+
0.42485782504081726,
|
| 977 |
+
0.37669673562049866,
|
| 978 |
+
0.4912723898887634,
|
| 979 |
+
0.22055673599243164,
|
| 980 |
+
0.627392053604126,
|
| 981 |
+
0.49359387159347534,
|
| 982 |
+
0.48515188694000244,
|
| 983 |
+
0.4972233772277832,
|
| 984 |
+
0.8575774431228638,
|
| 985 |
+
0.910191535949707,
|
| 986 |
+
0.8751074075698853,
|
| 987 |
+
0.8624600172042847,
|
| 988 |
+
0.8954752683639526,
|
| 989 |
+
0.5143798589706421,
|
| 990 |
+
0.4631466567516327,
|
| 991 |
+
0.4894852042198181,
|
| 992 |
+
0.6127458214759827,
|
| 993 |
+
0.5760451555252075,
|
| 994 |
+
0.6216609477996826,
|
| 995 |
+
0.5542522072792053,
|
| 996 |
+
0.6066001653671265,
|
| 997 |
+
0.4518378674983978,
|
| 998 |
+
0.5172532200813293,
|
| 999 |
+
0.5548712611198425,
|
| 1000 |
+
0.47210246324539185,
|
| 1001 |
+
0.480351984500885,
|
| 1002 |
+
0.5317205190658569,
|
| 1003 |
+
0.5237709283828735,
|
| 1004 |
+
0.5205705165863037,
|
| 1005 |
+
0.5198208093643188,
|
| 1006 |
+
0.5110467076301575,
|
| 1007 |
+
0.5243876576423645,
|
| 1008 |
+
0.6869658827781677,
|
| 1009 |
+
0.7886658310890198,
|
| 1010 |
+
0.8274509906768799,
|
| 1011 |
+
0.5666228532791138,
|
| 1012 |
+
0.5315422415733337,
|
| 1013 |
+
0.4891224503517151,
|
| 1014 |
+
0.4611716568470001,
|
| 1015 |
+
0.5529481768608093,
|
| 1016 |
+
0.6163869500160217,
|
| 1017 |
+
0.47675707936286926,
|
| 1018 |
+
0.5012125968933105,
|
| 1019 |
+
0.5132050514221191,
|
| 1020 |
+
0.5116791725158691,
|
| 1021 |
+
0.49521586298942566,
|
| 1022 |
+
0.4946640133857727,
|
| 1023 |
+
0.5137390494346619,
|
| 1024 |
+
0.4953595697879791,
|
| 1025 |
+
0.5362335443496704,
|
| 1026 |
+
0.7229145169258118,
|
| 1027 |
+
0.4241175949573517,
|
| 1028 |
+
0.3835858404636383,
|
| 1029 |
+
0.8711888790130615,
|
| 1030 |
+
0.8752027153968811,
|
| 1031 |
+
0.8618167638778687,
|
| 1032 |
+
0.4735512435436249,
|
| 1033 |
+
0.47097814083099365,
|
| 1034 |
+
0.4957125186920166,
|
| 1035 |
+
0.8091335296630859,
|
| 1036 |
+
0.8663551807403564,
|
| 1037 |
+
0.8784079551696777,
|
| 1038 |
+
0.5133549571037292,
|
| 1039 |
+
0.42201387882232666,
|
| 1040 |
+
0.4296361207962036,
|
| 1041 |
+
0.3151407241821289,
|
| 1042 |
+
0.5211864113807678,
|
| 1043 |
+
0.5909596085548401,
|
| 1044 |
+
0.524550199508667,
|
| 1045 |
+
0.7371199131011963,
|
| 1046 |
+
0.5599789619445801,
|
| 1047 |
+
0.04795292019844055,
|
| 1048 |
+
0.045255232602357864,
|
| 1049 |
+
0.04756557196378708,
|
| 1050 |
+
0.5445749759674072
|
| 1051 |
+
],
|
| 1052 |
+
"eval_math_runtime": 22.5981,
|
| 1053 |
+
"eval_math_samples_per_second": 4.425,
|
| 1054 |
+
"eval_math_steps_per_second": 0.089,
|
| 1055 |
+
"epoch": 1.75,
|
| 1056 |
+
"step": 100
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"train_runtime": 2129.8823,
|
| 1060 |
+
"train_samples_per_second": 1.502,
|
| 1061 |
+
"train_steps_per_second": 0.047,
|
| 1062 |
+
"total_flos": 0.0,
|
| 1063 |
+
"train_loss": 0.6600256007909775,
|
| 1064 |
+
"epoch": 1.75,
|
| 1065 |
+
"step": 100
|
| 1066 |
+
}
|
| 1067 |
+
]
|