Commit ·
54368b4
1
Parent(s): ddee128
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 +32 -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/llama-7b",
|
| 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:ad1a8b7a5d2c17c880db2f2c94b2a9651845820f8a84a24fb240a7d3494691e4
|
| 3 |
+
size 134295057
|
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 sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"model_max_length": 2048,
|
| 20 |
+
"pad_token": null,
|
| 21 |
+
"sp_model_kwargs": {},
|
| 22 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"__type": "AddedToken",
|
| 25 |
+
"content": "<unk>",
|
| 26 |
+
"lstrip": false,
|
| 27 |
+
"normalized": true,
|
| 28 |
+
"rstrip": false,
|
| 29 |
+
"single_word": false
|
| 30 |
+
},
|
| 31 |
+
"use_default_system_prompt": true
|
| 32 |
+
}
|
train_args.json
ADDED
|
@@ -0,0 +1,123 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"initial_model_dir": "models/llama-7b",
|
| 3 |
+
"distribution_id": "mmlu",
|
| 4 |
+
"date_trained": "10/10/2023 02:33:55",
|
| 5 |
+
"output_dir": "models/classify_lora/llama-7b-mmlu",
|
| 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": 0.0002,
|
| 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/llama-7b-mmlu/runs/Oct10_01-34-53_compute-permanent-node-975",
|
| 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": 1,
|
| 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-llama-7b-mmlu",
|
| 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_mmlu_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 sha256:912ad274682fd417914fd8d8cc895e6a5f52ecaebdc7e9684c46eb979e7a1684
|
| 3 |
+
size 5051
|
training_logs.json
ADDED
|
@@ -0,0 +1,1067 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
[
|
| 2 |
+
{
|
| 3 |
+
"loss": 0.7339,
|
| 4 |
+
"learning_rate": 0.0002,
|
| 5 |
+
"epoch": 0.02,
|
| 6 |
+
"step": 1
|
| 7 |
+
},
|
| 8 |
+
{
|
| 9 |
+
"loss": 0.7581,
|
| 10 |
+
"learning_rate": 0.0002,
|
| 11 |
+
"epoch": 0.04,
|
| 12 |
+
"step": 2
|
| 13 |
+
},
|
| 14 |
+
{
|
| 15 |
+
"loss": 0.7597,
|
| 16 |
+
"learning_rate": 0.0002,
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"step": 3
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"loss": 0.8837,
|
| 22 |
+
"learning_rate": 0.0002,
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
+
"step": 4
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"loss": 0.7708,
|
| 28 |
+
"learning_rate": 0.0002,
|
| 29 |
+
"epoch": 0.09,
|
| 30 |
+
"step": 5
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"loss": 0.6677,
|
| 34 |
+
"learning_rate": 0.0002,
|
| 35 |
+
"epoch": 0.11,
|
| 36 |
+
"step": 6
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"loss": 0.7334,
|
| 40 |
+
"learning_rate": 0.0002,
|
| 41 |
+
"epoch": 0.12,
|
| 42 |
+
"step": 7
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"loss": 0.6964,
|
| 46 |
+
"learning_rate": 0.0002,
|
| 47 |
+
"epoch": 0.14,
|
| 48 |
+
"step": 8
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"loss": 0.7324,
|
| 52 |
+
"learning_rate": 0.0002,
|
| 53 |
+
"epoch": 0.16,
|
| 54 |
+
"step": 9
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"loss": 0.7191,
|
| 58 |
+
"learning_rate": 0.0002,
|
| 59 |
+
"epoch": 0.18,
|
| 60 |
+
"step": 10
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"loss": 0.7533,
|
| 64 |
+
"learning_rate": 0.0002,
|
| 65 |
+
"epoch": 0.19,
|
| 66 |
+
"step": 11
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"loss": 0.7787,
|
| 70 |
+
"learning_rate": 0.0002,
|
| 71 |
+
"epoch": 0.21,
|
| 72 |
+
"step": 12
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"loss": 0.7004,
|
| 76 |
+
"learning_rate": 0.0002,
|
| 77 |
+
"epoch": 0.23,
|
| 78 |
+
"step": 13
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"loss": 0.685,
|
| 82 |
+
"learning_rate": 0.0002,
|
| 83 |
+
"epoch": 0.25,
|
| 84 |
+
"step": 14
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"loss": 0.6909,
|
| 88 |
+
"learning_rate": 0.0002,
|
| 89 |
+
"epoch": 0.26,
|
| 90 |
+
"step": 15
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"loss": 0.6928,
|
| 94 |
+
"learning_rate": 0.0002,
|
| 95 |
+
"epoch": 0.28,
|
| 96 |
+
"step": 16
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"loss": 0.7092,
|
| 100 |
+
"learning_rate": 0.0002,
|
| 101 |
+
"epoch": 0.3,
|
| 102 |
+
"step": 17
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"loss": 0.6884,
|
| 106 |
+
"learning_rate": 0.0002,
|
| 107 |
+
"epoch": 0.32,
|
| 108 |
+
"step": 18
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"loss": 0.7099,
|
| 112 |
+
"learning_rate": 0.0002,
|
| 113 |
+
"epoch": 0.33,
|
| 114 |
+
"step": 19
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"loss": 0.7124,
|
| 118 |
+
"learning_rate": 0.0002,
|
| 119 |
+
"epoch": 0.35,
|
| 120 |
+
"step": 20
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"loss": 0.692,
|
| 124 |
+
"learning_rate": 0.0002,
|
| 125 |
+
"epoch": 0.37,
|
| 126 |
+
"step": 21
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"loss": 0.6696,
|
| 130 |
+
"learning_rate": 0.0002,
|
| 131 |
+
"epoch": 0.39,
|
| 132 |
+
"step": 22
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"loss": 0.7233,
|
| 136 |
+
"learning_rate": 0.0002,
|
| 137 |
+
"epoch": 0.4,
|
| 138 |
+
"step": 23
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"loss": 0.7043,
|
| 142 |
+
"learning_rate": 0.0002,
|
| 143 |
+
"epoch": 0.42,
|
| 144 |
+
"step": 24
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"loss": 0.6885,
|
| 148 |
+
"learning_rate": 0.0002,
|
| 149 |
+
"epoch": 0.44,
|
| 150 |
+
"step": 25
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"eval_mmlu_loss": 0.6877026557922363,
|
| 154 |
+
"eval_mmlu_score": -0.24597446620464325,
|
| 155 |
+
"eval_mmlu_brier_score": 0.24597446620464325,
|
| 156 |
+
"eval_mmlu_average_probability": 0.5076507925987244,
|
| 157 |
+
"eval_mmlu_accuracy": 0.52,
|
| 158 |
+
"eval_mmlu_probabilities": [
|
| 159 |
+
0.5024531483650208,
|
| 160 |
+
0.49637264013290405,
|
| 161 |
+
0.42682787775993347,
|
| 162 |
+
0.5511913299560547,
|
| 163 |
+
0.516057014465332,
|
| 164 |
+
0.5302246809005737,
|
| 165 |
+
0.47325557470321655,
|
| 166 |
+
0.47100546956062317,
|
| 167 |
+
0.4857306480407715,
|
| 168 |
+
0.4879577159881592,
|
| 169 |
+
0.5080811977386475,
|
| 170 |
+
0.49308809638023376,
|
| 171 |
+
0.5444069504737854,
|
| 172 |
+
0.5699678659439087,
|
| 173 |
+
0.6137876510620117,
|
| 174 |
+
0.5039758682250977,
|
| 175 |
+
0.5088179707527161,
|
| 176 |
+
0.47646814584732056,
|
| 177 |
+
0.5660363435745239,
|
| 178 |
+
0.5961596965789795,
|
| 179 |
+
0.6173415184020996,
|
| 180 |
+
0.5275791883468628,
|
| 181 |
+
0.4939135015010834,
|
| 182 |
+
0.5678824782371521,
|
| 183 |
+
0.4228115677833557,
|
| 184 |
+
0.34861868619918823,
|
| 185 |
+
0.404059499502182,
|
| 186 |
+
0.5350965261459351,
|
| 187 |
+
0.6017758846282959,
|
| 188 |
+
0.5842959880828857,
|
| 189 |
+
0.5007165670394897,
|
| 190 |
+
0.5204458832740784,
|
| 191 |
+
0.46593177318573,
|
| 192 |
+
0.47197505831718445,
|
| 193 |
+
0.4855360984802246,
|
| 194 |
+
0.4904650151729584,
|
| 195 |
+
0.5280486941337585,
|
| 196 |
+
0.5117051005363464,
|
| 197 |
+
0.5350059866905212,
|
| 198 |
+
0.5049182772636414,
|
| 199 |
+
0.4832833409309387,
|
| 200 |
+
0.5799784064292908,
|
| 201 |
+
0.5069881677627563,
|
| 202 |
+
0.6448580026626587,
|
| 203 |
+
0.6689260005950928,
|
| 204 |
+
0.46061623096466064,
|
| 205 |
+
0.3911752700805664,
|
| 206 |
+
0.39503976702690125,
|
| 207 |
+
0.4661625623703003,
|
| 208 |
+
0.6794996857643127,
|
| 209 |
+
0.47264567017555237,
|
| 210 |
+
0.4458942115306854,
|
| 211 |
+
0.45081019401550293,
|
| 212 |
+
0.4047556221485138,
|
| 213 |
+
0.4076286554336548,
|
| 214 |
+
0.592082679271698,
|
| 215 |
+
0.534106433391571,
|
| 216 |
+
0.5288624167442322,
|
| 217 |
+
0.49373653531074524,
|
| 218 |
+
0.5440691709518433,
|
| 219 |
+
0.508944034576416,
|
| 220 |
+
0.5440325140953064,
|
| 221 |
+
0.5153629183769226,
|
| 222 |
+
0.466998428106308,
|
| 223 |
+
0.4557458758354187,
|
| 224 |
+
0.48426082730293274,
|
| 225 |
+
0.4218427240848541,
|
| 226 |
+
0.5727195739746094,
|
| 227 |
+
0.4905124604701996,
|
| 228 |
+
0.590722382068634,
|
| 229 |
+
0.5299772024154663,
|
| 230 |
+
0.5239810347557068,
|
| 231 |
+
0.600125253200531,
|
| 232 |
+
0.42955777049064636,
|
| 233 |
+
0.5403964519500732,
|
| 234 |
+
0.48015835881233215,
|
| 235 |
+
0.4788738489151001,
|
| 236 |
+
0.5453965663909912,
|
| 237 |
+
0.6047127842903137,
|
| 238 |
+
0.5861594676971436,
|
| 239 |
+
0.5148264169692993,
|
| 240 |
+
0.551675021648407,
|
| 241 |
+
0.5116910338401794,
|
| 242 |
+
0.5246034264564514,
|
| 243 |
+
0.49715280532836914,
|
| 244 |
+
0.48748254776000977,
|
| 245 |
+
0.44811931252479553,
|
| 246 |
+
0.4952526390552521,
|
| 247 |
+
0.4878782033920288,
|
| 248 |
+
0.4950997531414032,
|
| 249 |
+
0.4488215446472168,
|
| 250 |
+
0.531753659248352,
|
| 251 |
+
0.5307499766349792,
|
| 252 |
+
0.46273279190063477,
|
| 253 |
+
0.44232308864593506,
|
| 254 |
+
0.5162163972854614,
|
| 255 |
+
0.47405269742012024,
|
| 256 |
+
0.47172489762306213,
|
| 257 |
+
0.4590320885181427,
|
| 258 |
+
0.4523000419139862
|
| 259 |
+
],
|
| 260 |
+
"eval_mmlu_runtime": 36.2867,
|
| 261 |
+
"eval_mmlu_samples_per_second": 2.756,
|
| 262 |
+
"eval_mmlu_steps_per_second": 0.055,
|
| 263 |
+
"epoch": 0.44,
|
| 264 |
+
"step": 25
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"loss": 0.6873,
|
| 268 |
+
"learning_rate": 0.0002,
|
| 269 |
+
"epoch": 0.46,
|
| 270 |
+
"step": 26
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"loss": 0.6727,
|
| 274 |
+
"learning_rate": 0.0002,
|
| 275 |
+
"epoch": 0.47,
|
| 276 |
+
"step": 27
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"loss": 0.6513,
|
| 280 |
+
"learning_rate": 0.0002,
|
| 281 |
+
"epoch": 0.49,
|
| 282 |
+
"step": 28
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"loss": 0.6854,
|
| 286 |
+
"learning_rate": 0.0002,
|
| 287 |
+
"epoch": 0.51,
|
| 288 |
+
"step": 29
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"loss": 0.6944,
|
| 292 |
+
"learning_rate": 0.0002,
|
| 293 |
+
"epoch": 0.53,
|
| 294 |
+
"step": 30
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"loss": 0.6698,
|
| 298 |
+
"learning_rate": 0.0002,
|
| 299 |
+
"epoch": 0.54,
|
| 300 |
+
"step": 31
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"loss": 0.6873,
|
| 304 |
+
"learning_rate": 0.0002,
|
| 305 |
+
"epoch": 0.56,
|
| 306 |
+
"step": 32
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"loss": 0.7038,
|
| 310 |
+
"learning_rate": 0.0002,
|
| 311 |
+
"epoch": 0.58,
|
| 312 |
+
"step": 33
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"loss": 0.6371,
|
| 316 |
+
"learning_rate": 0.0002,
|
| 317 |
+
"epoch": 0.6,
|
| 318 |
+
"step": 34
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"loss": 0.6949,
|
| 322 |
+
"learning_rate": 0.0002,
|
| 323 |
+
"epoch": 0.61,
|
| 324 |
+
"step": 35
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"loss": 0.6806,
|
| 328 |
+
"learning_rate": 0.0002,
|
| 329 |
+
"epoch": 0.63,
|
| 330 |
+
"step": 36
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"loss": 0.6348,
|
| 334 |
+
"learning_rate": 0.0002,
|
| 335 |
+
"epoch": 0.65,
|
| 336 |
+
"step": 37
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"loss": 0.6525,
|
| 340 |
+
"learning_rate": 0.0002,
|
| 341 |
+
"epoch": 0.67,
|
| 342 |
+
"step": 38
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"loss": 0.6781,
|
| 346 |
+
"learning_rate": 0.0002,
|
| 347 |
+
"epoch": 0.68,
|
| 348 |
+
"step": 39
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"loss": 0.6672,
|
| 352 |
+
"learning_rate": 0.0002,
|
| 353 |
+
"epoch": 0.7,
|
| 354 |
+
"step": 40
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"loss": 0.6829,
|
| 358 |
+
"learning_rate": 0.0002,
|
| 359 |
+
"epoch": 0.72,
|
| 360 |
+
"step": 41
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"loss": 0.6725,
|
| 364 |
+
"learning_rate": 0.0002,
|
| 365 |
+
"epoch": 0.74,
|
| 366 |
+
"step": 42
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"loss": 0.6299,
|
| 370 |
+
"learning_rate": 0.0002,
|
| 371 |
+
"epoch": 0.75,
|
| 372 |
+
"step": 43
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"loss": 0.679,
|
| 376 |
+
"learning_rate": 0.0002,
|
| 377 |
+
"epoch": 0.77,
|
| 378 |
+
"step": 44
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"loss": 0.6444,
|
| 382 |
+
"learning_rate": 0.0002,
|
| 383 |
+
"epoch": 0.79,
|
| 384 |
+
"step": 45
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"loss": 0.5872,
|
| 388 |
+
"learning_rate": 0.0002,
|
| 389 |
+
"epoch": 0.81,
|
| 390 |
+
"step": 46
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"loss": 0.5675,
|
| 394 |
+
"learning_rate": 0.0002,
|
| 395 |
+
"epoch": 0.82,
|
| 396 |
+
"step": 47
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"loss": 0.6037,
|
| 400 |
+
"learning_rate": 0.0002,
|
| 401 |
+
"epoch": 0.84,
|
| 402 |
+
"step": 48
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"loss": 0.6257,
|
| 406 |
+
"learning_rate": 0.0002,
|
| 407 |
+
"epoch": 0.86,
|
| 408 |
+
"step": 49
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"loss": 0.6217,
|
| 412 |
+
"learning_rate": 0.0002,
|
| 413 |
+
"epoch": 0.88,
|
| 414 |
+
"step": 50
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"eval_mmlu_loss": 0.6811871528625488,
|
| 418 |
+
"eval_mmlu_score": -0.2452215701341629,
|
| 419 |
+
"eval_mmlu_brier_score": 0.2452215701341629,
|
| 420 |
+
"eval_mmlu_average_probability": 0.5182498693466187,
|
| 421 |
+
"eval_mmlu_accuracy": 0.56,
|
| 422 |
+
"eval_mmlu_probabilities": [
|
| 423 |
+
0.5035430788993835,
|
| 424 |
+
0.4767981767654419,
|
| 425 |
+
0.5682129859924316,
|
| 426 |
+
0.5918246507644653,
|
| 427 |
+
0.5153557658195496,
|
| 428 |
+
0.5415547490119934,
|
| 429 |
+
0.4029051959514618,
|
| 430 |
+
0.450078547000885,
|
| 431 |
+
0.5172317028045654,
|
| 432 |
+
0.6108822822570801,
|
| 433 |
+
0.4666701555252075,
|
| 434 |
+
0.4877246916294098,
|
| 435 |
+
0.4729712903499603,
|
| 436 |
+
0.4151947796344757,
|
| 437 |
+
0.46747589111328125,
|
| 438 |
+
0.5594338178634644,
|
| 439 |
+
0.5088744759559631,
|
| 440 |
+
0.5103889107704163,
|
| 441 |
+
0.5036099553108215,
|
| 442 |
+
0.584050714969635,
|
| 443 |
+
0.6694682836532593,
|
| 444 |
+
0.5144792795181274,
|
| 445 |
+
0.5978407859802246,
|
| 446 |
+
0.6020472645759583,
|
| 447 |
+
0.4669976532459259,
|
| 448 |
+
0.5301037430763245,
|
| 449 |
+
0.49586060643196106,
|
| 450 |
+
0.5453556776046753,
|
| 451 |
+
0.8031986355781555,
|
| 452 |
+
0.7073482275009155,
|
| 453 |
+
0.6679326891899109,
|
| 454 |
+
0.4658517837524414,
|
| 455 |
+
0.37422868609428406,
|
| 456 |
+
0.37934908270835876,
|
| 457 |
+
0.4810079038143158,
|
| 458 |
+
0.6123296618461609,
|
| 459 |
+
0.4183340072631836,
|
| 460 |
+
0.5893468856811523,
|
| 461 |
+
0.4664745032787323,
|
| 462 |
+
0.41369232535362244,
|
| 463 |
+
0.414699912071228,
|
| 464 |
+
0.43868860602378845,
|
| 465 |
+
0.5881794095039368,
|
| 466 |
+
0.6734998226165771,
|
| 467 |
+
0.7834814786911011,
|
| 468 |
+
0.34748634696006775,
|
| 469 |
+
0.5293741226196289,
|
| 470 |
+
0.36080241203308105,
|
| 471 |
+
0.42220059037208557,
|
| 472 |
+
0.7680569887161255,
|
| 473 |
+
0.4100235402584076,
|
| 474 |
+
0.7189415097236633,
|
| 475 |
+
0.6853945851325989,
|
| 476 |
+
0.5016038417816162,
|
| 477 |
+
0.39125701785087585,
|
| 478 |
+
0.7705144286155701,
|
| 479 |
+
0.6471617817878723,
|
| 480 |
+
0.3062707483768463,
|
| 481 |
+
0.386313259601593,
|
| 482 |
+
0.32649141550064087,
|
| 483 |
+
0.47263073921203613,
|
| 484 |
+
0.44739583134651184,
|
| 485 |
+
0.5802536010742188,
|
| 486 |
+
0.4064495265483856,
|
| 487 |
+
0.4570828974246979,
|
| 488 |
+
0.4176960289478302,
|
| 489 |
+
0.28201282024383545,
|
| 490 |
+
0.6262605786323547,
|
| 491 |
+
0.617345929145813,
|
| 492 |
+
0.41387128829956055,
|
| 493 |
+
0.5746867060661316,
|
| 494 |
+
0.5823390483856201,
|
| 495 |
+
0.3908570408821106,
|
| 496 |
+
0.31184154748916626,
|
| 497 |
+
0.515109658241272,
|
| 498 |
+
0.43107518553733826,
|
| 499 |
+
0.4796065390110016,
|
| 500 |
+
0.7087035775184631,
|
| 501 |
+
0.6558853983879089,
|
| 502 |
+
0.5968166589736938,
|
| 503 |
+
0.47654810547828674,
|
| 504 |
+
0.5344870686531067,
|
| 505 |
+
0.6017045974731445,
|
| 506 |
+
0.5343926548957825,
|
| 507 |
+
0.5024454593658447,
|
| 508 |
+
0.5379210710525513,
|
| 509 |
+
0.42640140652656555,
|
| 510 |
+
0.6160038709640503,
|
| 511 |
+
0.5520896315574646,
|
| 512 |
+
0.4460159242153168,
|
| 513 |
+
0.5341697931289673,
|
| 514 |
+
0.6633436679840088,
|
| 515 |
+
0.536496102809906,
|
| 516 |
+
0.6880946159362793,
|
| 517 |
+
0.6644284129142761,
|
| 518 |
+
0.5711739659309387,
|
| 519 |
+
0.4048452377319336,
|
| 520 |
+
0.2511211037635803,
|
| 521 |
+
0.38734522461891174,
|
| 522 |
+
0.5015671253204346
|
| 523 |
+
],
|
| 524 |
+
"eval_mmlu_runtime": 36.3017,
|
| 525 |
+
"eval_mmlu_samples_per_second": 2.755,
|
| 526 |
+
"eval_mmlu_steps_per_second": 0.055,
|
| 527 |
+
"epoch": 0.88,
|
| 528 |
+
"step": 50
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"loss": 0.6395,
|
| 532 |
+
"learning_rate": 0.0002,
|
| 533 |
+
"epoch": 0.89,
|
| 534 |
+
"step": 51
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"loss": 0.5624,
|
| 538 |
+
"learning_rate": 0.0002,
|
| 539 |
+
"epoch": 0.91,
|
| 540 |
+
"step": 52
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"loss": 0.6642,
|
| 544 |
+
"learning_rate": 0.0002,
|
| 545 |
+
"epoch": 0.93,
|
| 546 |
+
"step": 53
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"loss": 0.6446,
|
| 550 |
+
"learning_rate": 0.0002,
|
| 551 |
+
"epoch": 0.95,
|
| 552 |
+
"step": 54
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"loss": 0.5786,
|
| 556 |
+
"learning_rate": 0.0002,
|
| 557 |
+
"epoch": 0.96,
|
| 558 |
+
"step": 55
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"loss": 0.6207,
|
| 562 |
+
"learning_rate": 0.0002,
|
| 563 |
+
"epoch": 0.98,
|
| 564 |
+
"step": 56
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"loss": 0.5365,
|
| 568 |
+
"learning_rate": 0.0002,
|
| 569 |
+
"epoch": 1.0,
|
| 570 |
+
"step": 57
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"loss": 0.5721,
|
| 574 |
+
"learning_rate": 0.0002,
|
| 575 |
+
"epoch": 1.02,
|
| 576 |
+
"step": 58
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"loss": 0.5258,
|
| 580 |
+
"learning_rate": 0.0002,
|
| 581 |
+
"epoch": 1.04,
|
| 582 |
+
"step": 59
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"loss": 0.5292,
|
| 586 |
+
"learning_rate": 0.0002,
|
| 587 |
+
"epoch": 1.05,
|
| 588 |
+
"step": 60
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"loss": 0.5543,
|
| 592 |
+
"learning_rate": 0.0002,
|
| 593 |
+
"epoch": 1.07,
|
| 594 |
+
"step": 61
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"loss": 0.539,
|
| 598 |
+
"learning_rate": 0.0002,
|
| 599 |
+
"epoch": 1.09,
|
| 600 |
+
"step": 62
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"loss": 0.5608,
|
| 604 |
+
"learning_rate": 0.0002,
|
| 605 |
+
"epoch": 1.11,
|
| 606 |
+
"step": 63
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"loss": 0.4825,
|
| 610 |
+
"learning_rate": 0.0002,
|
| 611 |
+
"epoch": 1.12,
|
| 612 |
+
"step": 64
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"loss": 0.4465,
|
| 616 |
+
"learning_rate": 0.0002,
|
| 617 |
+
"epoch": 1.14,
|
| 618 |
+
"step": 65
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"loss": 0.3743,
|
| 622 |
+
"learning_rate": 0.0002,
|
| 623 |
+
"epoch": 1.16,
|
| 624 |
+
"step": 66
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"loss": 0.5076,
|
| 628 |
+
"learning_rate": 0.0002,
|
| 629 |
+
"epoch": 1.18,
|
| 630 |
+
"step": 67
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"loss": 0.435,
|
| 634 |
+
"learning_rate": 0.0002,
|
| 635 |
+
"epoch": 1.19,
|
| 636 |
+
"step": 68
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"loss": 0.3346,
|
| 640 |
+
"learning_rate": 0.0002,
|
| 641 |
+
"epoch": 1.21,
|
| 642 |
+
"step": 69
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"loss": 0.5868,
|
| 646 |
+
"learning_rate": 0.0002,
|
| 647 |
+
"epoch": 1.23,
|
| 648 |
+
"step": 70
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"loss": 0.3487,
|
| 652 |
+
"learning_rate": 0.0002,
|
| 653 |
+
"epoch": 1.25,
|
| 654 |
+
"step": 71
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"loss": 0.3667,
|
| 658 |
+
"learning_rate": 0.0002,
|
| 659 |
+
"epoch": 1.26,
|
| 660 |
+
"step": 72
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"loss": 0.5619,
|
| 664 |
+
"learning_rate": 0.0002,
|
| 665 |
+
"epoch": 1.28,
|
| 666 |
+
"step": 73
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"loss": 0.3931,
|
| 670 |
+
"learning_rate": 0.0002,
|
| 671 |
+
"epoch": 1.3,
|
| 672 |
+
"step": 74
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"loss": 0.3349,
|
| 676 |
+
"learning_rate": 0.0002,
|
| 677 |
+
"epoch": 1.32,
|
| 678 |
+
"step": 75
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"eval_mmlu_loss": 0.833143413066864,
|
| 682 |
+
"eval_mmlu_score": -0.2877655327320099,
|
| 683 |
+
"eval_mmlu_brier_score": 0.2877655327320099,
|
| 684 |
+
"eval_mmlu_average_probability": 0.5490172505378723,
|
| 685 |
+
"eval_mmlu_accuracy": 0.54,
|
| 686 |
+
"eval_mmlu_probabilities": [
|
| 687 |
+
0.1580410748720169,
|
| 688 |
+
0.43874940276145935,
|
| 689 |
+
0.6015041470527649,
|
| 690 |
+
0.5842825770378113,
|
| 691 |
+
0.4584586024284363,
|
| 692 |
+
0.4909316897392273,
|
| 693 |
+
0.11258751899003983,
|
| 694 |
+
0.38813868165016174,
|
| 695 |
+
0.23412899672985077,
|
| 696 |
+
0.8689023852348328,
|
| 697 |
+
0.3216671347618103,
|
| 698 |
+
0.4708637595176697,
|
| 699 |
+
0.6932336688041687,
|
| 700 |
+
0.9213990569114685,
|
| 701 |
+
0.44838589429855347,
|
| 702 |
+
0.6026622653007507,
|
| 703 |
+
0.5027726888656616,
|
| 704 |
+
0.612128496170044,
|
| 705 |
+
0.06997516006231308,
|
| 706 |
+
0.4677487909793854,
|
| 707 |
+
0.35034096240997314,
|
| 708 |
+
0.8762949705123901,
|
| 709 |
+
0.7732230424880981,
|
| 710 |
+
0.8989255428314209,
|
| 711 |
+
0.40452077984809875,
|
| 712 |
+
0.48415499925613403,
|
| 713 |
+
0.374180406332016,
|
| 714 |
+
0.9165855050086975,
|
| 715 |
+
0.9918805956840515,
|
| 716 |
+
0.9852085113525391,
|
| 717 |
+
0.8175538778305054,
|
| 718 |
+
0.09847690165042877,
|
| 719 |
+
0.26064637303352356,
|
| 720 |
+
0.46926015615463257,
|
| 721 |
+
0.8475375771522522,
|
| 722 |
+
0.7741795778274536,
|
| 723 |
+
0.26930543780326843,
|
| 724 |
+
0.4011392295360565,
|
| 725 |
+
0.07169023901224136,
|
| 726 |
+
0.13525348901748657,
|
| 727 |
+
0.15383993089199066,
|
| 728 |
+
0.534102201461792,
|
| 729 |
+
0.8887997269630432,
|
| 730 |
+
0.983967125415802,
|
| 731 |
+
0.9906513690948486,
|
| 732 |
+
0.13722844421863556,
|
| 733 |
+
0.08100348711013794,
|
| 734 |
+
0.050758879631757736,
|
| 735 |
+
0.8930805921554565,
|
| 736 |
+
0.7616145014762878,
|
| 737 |
+
0.5024493932723999,
|
| 738 |
+
0.31655770540237427,
|
| 739 |
+
0.7933811545372009,
|
| 740 |
+
0.53852778673172,
|
| 741 |
+
0.9475235342979431,
|
| 742 |
+
0.9588028192520142,
|
| 743 |
+
0.9672081470489502,
|
| 744 |
+
0.23299279808998108,
|
| 745 |
+
0.5359262824058533,
|
| 746 |
+
0.39278143644332886,
|
| 747 |
+
0.8811087012290955,
|
| 748 |
+
0.89657062292099,
|
| 749 |
+
0.9659839868545532,
|
| 750 |
+
0.6707325577735901,
|
| 751 |
+
0.7787579894065857,
|
| 752 |
+
0.5808053612709045,
|
| 753 |
+
0.9807056784629822,
|
| 754 |
+
0.9567878842353821,
|
| 755 |
+
0.9984741806983948,
|
| 756 |
+
0.025270536541938782,
|
| 757 |
+
0.5340918898582458,
|
| 758 |
+
0.2554855942726135,
|
| 759 |
+
0.018762007355690002,
|
| 760 |
+
0.15030156075954437,
|
| 761 |
+
0.516846239566803,
|
| 762 |
+
0.2843777537345886,
|
| 763 |
+
0.410349577665329,
|
| 764 |
+
0.6759911179542542,
|
| 765 |
+
0.7385581135749817,
|
| 766 |
+
0.8582839965820312,
|
| 767 |
+
0.18342167139053345,
|
| 768 |
+
0.3932557702064514,
|
| 769 |
+
0.9066393375396729,
|
| 770 |
+
0.41418227553367615,
|
| 771 |
+
0.6529518961906433,
|
| 772 |
+
0.5801039934158325,
|
| 773 |
+
0.10069625079631805,
|
| 774 |
+
0.6211609244346619,
|
| 775 |
+
0.3417372703552246,
|
| 776 |
+
0.3543941378593445,
|
| 777 |
+
0.6216508746147156,
|
| 778 |
+
0.8545634150505066,
|
| 779 |
+
0.574700653553009,
|
| 780 |
+
0.9054378271102905,
|
| 781 |
+
0.8045241832733154,
|
| 782 |
+
0.7795861959457397,
|
| 783 |
+
0.4488513469696045,
|
| 784 |
+
0.12007670104503632,
|
| 785 |
+
0.40916627645492554,
|
| 786 |
+
0.34826213121414185
|
| 787 |
+
],
|
| 788 |
+
"eval_mmlu_runtime": 36.2692,
|
| 789 |
+
"eval_mmlu_samples_per_second": 2.757,
|
| 790 |
+
"eval_mmlu_steps_per_second": 0.055,
|
| 791 |
+
"epoch": 1.32,
|
| 792 |
+
"step": 75
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"loss": 0.473,
|
| 796 |
+
"learning_rate": 0.0002,
|
| 797 |
+
"epoch": 1.33,
|
| 798 |
+
"step": 76
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"loss": 0.4736,
|
| 802 |
+
"learning_rate": 0.0002,
|
| 803 |
+
"epoch": 1.35,
|
| 804 |
+
"step": 77
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"loss": 0.4351,
|
| 808 |
+
"learning_rate": 0.0002,
|
| 809 |
+
"epoch": 1.37,
|
| 810 |
+
"step": 78
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"loss": 0.2961,
|
| 814 |
+
"learning_rate": 0.0002,
|
| 815 |
+
"epoch": 1.39,
|
| 816 |
+
"step": 79
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"loss": 0.473,
|
| 820 |
+
"learning_rate": 0.0002,
|
| 821 |
+
"epoch": 1.4,
|
| 822 |
+
"step": 80
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"loss": 0.3997,
|
| 826 |
+
"learning_rate": 0.0002,
|
| 827 |
+
"epoch": 1.42,
|
| 828 |
+
"step": 81
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"loss": 0.3277,
|
| 832 |
+
"learning_rate": 0.0002,
|
| 833 |
+
"epoch": 1.44,
|
| 834 |
+
"step": 82
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"loss": 0.5077,
|
| 838 |
+
"learning_rate": 0.0002,
|
| 839 |
+
"epoch": 1.46,
|
| 840 |
+
"step": 83
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"loss": 0.3042,
|
| 844 |
+
"learning_rate": 0.0002,
|
| 845 |
+
"epoch": 1.47,
|
| 846 |
+
"step": 84
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"loss": 0.381,
|
| 850 |
+
"learning_rate": 0.0002,
|
| 851 |
+
"epoch": 1.49,
|
| 852 |
+
"step": 85
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"loss": 0.2834,
|
| 856 |
+
"learning_rate": 0.0002,
|
| 857 |
+
"epoch": 1.51,
|
| 858 |
+
"step": 86
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"loss": 0.3241,
|
| 862 |
+
"learning_rate": 0.0002,
|
| 863 |
+
"epoch": 1.53,
|
| 864 |
+
"step": 87
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"loss": 0.4027,
|
| 868 |
+
"learning_rate": 0.0002,
|
| 869 |
+
"epoch": 1.54,
|
| 870 |
+
"step": 88
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"loss": 0.3334,
|
| 874 |
+
"learning_rate": 0.0002,
|
| 875 |
+
"epoch": 1.56,
|
| 876 |
+
"step": 89
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"loss": 0.3422,
|
| 880 |
+
"learning_rate": 0.0002,
|
| 881 |
+
"epoch": 1.58,
|
| 882 |
+
"step": 90
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"loss": 0.368,
|
| 886 |
+
"learning_rate": 0.0002,
|
| 887 |
+
"epoch": 1.6,
|
| 888 |
+
"step": 91
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"loss": 0.3376,
|
| 892 |
+
"learning_rate": 0.0002,
|
| 893 |
+
"epoch": 1.61,
|
| 894 |
+
"step": 92
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"loss": 0.5442,
|
| 898 |
+
"learning_rate": 0.0002,
|
| 899 |
+
"epoch": 1.63,
|
| 900 |
+
"step": 93
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"loss": 0.3687,
|
| 904 |
+
"learning_rate": 0.0002,
|
| 905 |
+
"epoch": 1.65,
|
| 906 |
+
"step": 94
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"loss": 0.3144,
|
| 910 |
+
"learning_rate": 0.0002,
|
| 911 |
+
"epoch": 1.67,
|
| 912 |
+
"step": 95
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"loss": 0.4151,
|
| 916 |
+
"learning_rate": 0.0002,
|
| 917 |
+
"epoch": 1.68,
|
| 918 |
+
"step": 96
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"loss": 0.2863,
|
| 922 |
+
"learning_rate": 0.0002,
|
| 923 |
+
"epoch": 1.7,
|
| 924 |
+
"step": 97
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"loss": 0.464,
|
| 928 |
+
"learning_rate": 0.0002,
|
| 929 |
+
"epoch": 1.72,
|
| 930 |
+
"step": 98
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"loss": 0.2873,
|
| 934 |
+
"learning_rate": 0.0002,
|
| 935 |
+
"epoch": 1.74,
|
| 936 |
+
"step": 99
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"loss": 0.3988,
|
| 940 |
+
"learning_rate": 0.0002,
|
| 941 |
+
"epoch": 1.75,
|
| 942 |
+
"step": 100
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"eval_mmlu_loss": 0.8989719152450562,
|
| 946 |
+
"eval_mmlu_score": -0.3084475100040436,
|
| 947 |
+
"eval_mmlu_brier_score": 0.3084475100040436,
|
| 948 |
+
"eval_mmlu_average_probability": 0.5211965441703796,
|
| 949 |
+
"eval_mmlu_accuracy": 0.49,
|
| 950 |
+
"eval_mmlu_probabilities": [
|
| 951 |
+
0.4071880877017975,
|
| 952 |
+
0.45392853021621704,
|
| 953 |
+
0.7654134035110474,
|
| 954 |
+
0.8478649854660034,
|
| 955 |
+
0.2505261301994324,
|
| 956 |
+
0.4063345193862915,
|
| 957 |
+
0.27478206157684326,
|
| 958 |
+
0.3481687009334564,
|
| 959 |
+
0.20378506183624268,
|
| 960 |
+
0.7234545946121216,
|
| 961 |
+
0.4079451560974121,
|
| 962 |
+
0.46001574397087097,
|
| 963 |
+
0.28416189551353455,
|
| 964 |
+
0.4730168879032135,
|
| 965 |
+
0.4832534193992615,
|
| 966 |
+
0.6298182010650635,
|
| 967 |
+
0.5491191148757935,
|
| 968 |
+
0.7803859114646912,
|
| 969 |
+
0.48246490955352783,
|
| 970 |
+
0.7818248271942139,
|
| 971 |
+
0.8155978322029114,
|
| 972 |
+
0.7227077484130859,
|
| 973 |
+
0.5422055125236511,
|
| 974 |
+
0.3796870708465576,
|
| 975 |
+
0.0472140833735466,
|
| 976 |
+
0.13505573570728302,
|
| 977 |
+
0.5995903015136719,
|
| 978 |
+
0.7262413501739502,
|
| 979 |
+
0.9919582605361938,
|
| 980 |
+
0.9893267154693604,
|
| 981 |
+
0.673717737197876,
|
| 982 |
+
0.09111680835485458,
|
| 983 |
+
0.062387943267822266,
|
| 984 |
+
0.18485485017299652,
|
| 985 |
+
0.831606924533844,
|
| 986 |
+
0.9187701940536499,
|
| 987 |
+
0.2769331932067871,
|
| 988 |
+
0.7295668125152588,
|
| 989 |
+
0.25232887268066406,
|
| 990 |
+
0.1371425986289978,
|
| 991 |
+
0.11290226131677628,
|
| 992 |
+
0.42969319224357605,
|
| 993 |
+
0.8279014825820923,
|
| 994 |
+
0.833659291267395,
|
| 995 |
+
0.995989978313446,
|
| 996 |
+
0.07782114297151566,
|
| 997 |
+
0.29245418310165405,
|
| 998 |
+
0.1428614854812622,
|
| 999 |
+
0.12554965913295746,
|
| 1000 |
+
0.885955810546875,
|
| 1001 |
+
0.21254442632198334,
|
| 1002 |
+
0.8990002274513245,
|
| 1003 |
+
0.7997588515281677,
|
| 1004 |
+
0.2548746168613434,
|
| 1005 |
+
0.5323441028594971,
|
| 1006 |
+
0.9922589659690857,
|
| 1007 |
+
0.7346373200416565,
|
| 1008 |
+
0.08055279403924942,
|
| 1009 |
+
0.27386149764060974,
|
| 1010 |
+
0.34895414113998413,
|
| 1011 |
+
0.7386496067047119,
|
| 1012 |
+
0.7762178182601929,
|
| 1013 |
+
0.7541271448135376,
|
| 1014 |
+
0.5595319271087646,
|
| 1015 |
+
0.5181352496147156,
|
| 1016 |
+
0.8376888036727905,
|
| 1017 |
+
0.228178009390831,
|
| 1018 |
+
0.9345813989639282,
|
| 1019 |
+
0.9527415037155151,
|
| 1020 |
+
0.03981766849756241,
|
| 1021 |
+
0.6531015038490295,
|
| 1022 |
+
0.7840538620948792,
|
| 1023 |
+
0.07283812761306763,
|
| 1024 |
+
0.06463921070098877,
|
| 1025 |
+
0.2090235948562622,
|
| 1026 |
+
0.380914568901062,
|
| 1027 |
+
0.4863632321357727,
|
| 1028 |
+
0.803974449634552,
|
| 1029 |
+
0.7102047204971313,
|
| 1030 |
+
0.8492692708969116,
|
| 1031 |
+
0.2411557286977768,
|
| 1032 |
+
0.4371771812438965,
|
| 1033 |
+
0.8251751065254211,
|
| 1034 |
+
0.2947361469268799,
|
| 1035 |
+
0.5158799886703491,
|
| 1036 |
+
0.4882834851741791,
|
| 1037 |
+
0.17904901504516602,
|
| 1038 |
+
0.4950083792209625,
|
| 1039 |
+
0.3899526596069336,
|
| 1040 |
+
0.30322733521461487,
|
| 1041 |
+
0.7750067710876465,
|
| 1042 |
+
0.9384438395500183,
|
| 1043 |
+
0.2649548649787903,
|
| 1044 |
+
0.956122636795044,
|
| 1045 |
+
0.8255246877670288,
|
| 1046 |
+
0.7141478657722473,
|
| 1047 |
+
0.44438719749450684,
|
| 1048 |
+
0.6701680421829224,
|
| 1049 |
+
0.599485456943512,
|
| 1050 |
+
0.43267613649368286
|
| 1051 |
+
],
|
| 1052 |
+
"eval_mmlu_runtime": 36.2731,
|
| 1053 |
+
"eval_mmlu_samples_per_second": 2.757,
|
| 1054 |
+
"eval_mmlu_steps_per_second": 0.055,
|
| 1055 |
+
"epoch": 1.75,
|
| 1056 |
+
"step": 100
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"train_runtime": 3535.9045,
|
| 1060 |
+
"train_samples_per_second": 0.905,
|
| 1061 |
+
"train_steps_per_second": 0.028,
|
| 1062 |
+
"total_flos": 0.0,
|
| 1063 |
+
"train_loss": 0.5670726171135902,
|
| 1064 |
+
"epoch": 1.75,
|
| 1065 |
+
"step": 100
|
| 1066 |
+
}
|
| 1067 |
+
]
|