Hugging Face's logo

andrk9
/
testLlama2

PEFT
llama
8-bit precision
bitsandbytes
Model card Files
xet
Community
andrk9 commited on
Commit
d3a0cb8
·
1 Parent(s): 658796c

Upload folder using huggingface_hub

Browse files
Files changed (45) hide show
  1. README.md +99 -0
  2. adapter_config.json +21 -0
  3. adapter_model.bin +3 -0
  4. checkpoint-730/README.md +21 -0
  5. checkpoint-730/adapter_config.json +21 -0
  6. checkpoint-730/adapter_model.bin +3 -0
  7. checkpoint-730/optimizer.pt +3 -0
  8. checkpoint-730/rng_state.pth +3 -0
  9. checkpoint-730/scheduler.pt +3 -0
  10. checkpoint-730/trainer_state.json +2063 -0
  11. checkpoint-730/training_args.bin +3 -0
  12. checkpoint-740/README.md +21 -0
  13. checkpoint-740/adapter_config.json +21 -0
  14. checkpoint-740/adapter_model.bin +3 -0
  15. checkpoint-740/optimizer.pt +3 -0
  16. checkpoint-740/rng_state.pth +3 -0
  17. checkpoint-740/scheduler.pt +3 -0
  18. checkpoint-740/trainer_state.json +2091 -0
  19. checkpoint-740/training_args.bin +3 -0
  20. checkpoint-750/README.md +21 -0
  21. checkpoint-750/adapter_config.json +21 -0
  22. checkpoint-750/adapter_model.bin +3 -0
  23. checkpoint-750/optimizer.pt +3 -0
  24. checkpoint-750/rng_state.pth +3 -0
  25. checkpoint-750/scheduler.pt +3 -0
  26. checkpoint-750/trainer_state.json +2119 -0
  27. checkpoint-750/training_args.bin +3 -0
  28. checkpoint-760/README.md +21 -0
  29. checkpoint-760/adapter_config.json +21 -0
  30. checkpoint-760/adapter_model.bin +3 -0
  31. checkpoint-760/optimizer.pt +3 -0
  32. checkpoint-760/rng_state.pth +3 -0
  33. checkpoint-760/scheduler.pt +3 -0
  34. checkpoint-760/trainer_state.json +2147 -0
  35. checkpoint-760/training_args.bin +3 -0
  36. checkpoint-770/README.md +21 -0
  37. checkpoint-770/adapter_config.json +21 -0
  38. checkpoint-770/adapter_model.bin +3 -0
  39. checkpoint-770/optimizer.pt +3 -0
  40. checkpoint-770/rng_state.pth +3 -0
  41. checkpoint-770/scheduler.pt +3 -0
  42. checkpoint-770/trainer_state.json +2175 -0
  43. checkpoint-770/training_args.bin +3 -0
  44. config.json +37 -0
  45. peft_config.json +23 -0
README.md ADDED
@@ -0,0 +1,99 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: True
10
+ - load_in_4bit: False
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: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+
19
+ The following `bitsandbytes` quantization config was used during training:
20
+ - quant_method: bitsandbytes
21
+ - load_in_8bit: True
22
+ - load_in_4bit: False
23
+ - llm_int8_threshold: 6.0
24
+ - llm_int8_skip_modules: None
25
+ - llm_int8_enable_fp32_cpu_offload: False
26
+ - llm_int8_has_fp16_weight: False
27
+ - bnb_4bit_quant_type: fp4
28
+ - bnb_4bit_use_double_quant: False
29
+ - bnb_4bit_compute_dtype: float32
30
+
31
+ The following `bitsandbytes` quantization config was used during training:
32
+ - quant_method: bitsandbytes
33
+ - load_in_8bit: True
34
+ - load_in_4bit: False
35
+ - llm_int8_threshold: 6.0
36
+ - llm_int8_skip_modules: None
37
+ - llm_int8_enable_fp32_cpu_offload: False
38
+ - llm_int8_has_fp16_weight: False
39
+ - bnb_4bit_quant_type: fp4
40
+ - bnb_4bit_use_double_quant: False
41
+ - bnb_4bit_compute_dtype: float32
42
+
43
+ The following `bitsandbytes` quantization config was used during training:
44
+ - quant_method: bitsandbytes
45
+ - load_in_8bit: True
46
+ - load_in_4bit: False
47
+ - llm_int8_threshold: 6.0
48
+ - llm_int8_skip_modules: None
49
+ - llm_int8_enable_fp32_cpu_offload: False
50
+ - llm_int8_has_fp16_weight: False
51
+ - bnb_4bit_quant_type: fp4
52
+ - bnb_4bit_use_double_quant: False
53
+ - bnb_4bit_compute_dtype: float32
54
+
55
+ The following `bitsandbytes` quantization config was used during training:
56
+ - quant_method: bitsandbytes
57
+ - load_in_8bit: True
58
+ - load_in_4bit: False
59
+ - llm_int8_threshold: 6.0
60
+ - llm_int8_skip_modules: None
61
+ - llm_int8_enable_fp32_cpu_offload: False
62
+ - llm_int8_has_fp16_weight: False
63
+ - bnb_4bit_quant_type: fp4
64
+ - bnb_4bit_use_double_quant: False
65
+ - bnb_4bit_compute_dtype: float32
66
+
67
+ The following `bitsandbytes` quantization config was used during training:
68
+ - quant_method: bitsandbytes
69
+ - load_in_8bit: True
70
+ - load_in_4bit: False
71
+ - llm_int8_threshold: 6.0
72
+ - llm_int8_skip_modules: None
73
+ - llm_int8_enable_fp32_cpu_offload: False
74
+ - llm_int8_has_fp16_weight: False
75
+ - bnb_4bit_quant_type: fp4
76
+ - bnb_4bit_use_double_quant: False
77
+ - bnb_4bit_compute_dtype: float32
78
+
79
+ The following `bitsandbytes` quantization config was used during training:
80
+ - quant_method: bitsandbytes
81
+ - load_in_8bit: True
82
+ - load_in_4bit: False
83
+ - llm_int8_threshold: 6.0
84
+ - llm_int8_skip_modules: None
85
+ - llm_int8_enable_fp32_cpu_offload: False
86
+ - llm_int8_has_fp16_weight: False
87
+ - bnb_4bit_quant_type: fp4
88
+ - bnb_4bit_use_double_quant: False
89
+ - bnb_4bit_compute_dtype: float32
90
+ ### Framework versions
91
+
92
+ - PEFT 0.5.0
93
+ - PEFT 0.5.0
94
+ - PEFT 0.5.0
95
+ - PEFT 0.5.0
96
+ - PEFT 0.5.0
97
+ - PEFT 0.5.0
98
+
99
+ - PEFT 0.5.0
adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
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": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c49461f6486069a6430d2f9ef32aa1fa24fe7eaf8b85f00acb6ae38ef6eb04c4
3
+ size 16822989
checkpoint-730/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: True
10
+ - load_in_4bit: False
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: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.5.0
checkpoint-730/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
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": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-730/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31d8320deb6b9c07c250c4e3c847260bfa49b771a79e3bb78ee06dd3ea84ae98
3
+ size 16822989
checkpoint-730/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77d4ef7c72c59726511a515c3a92aad272735c8f3a5a4bd2a7f877b3ab238b30
3
+ size 33665221
checkpoint-730/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d352df4a375eebfd23ed0d314f64a2d9383d67d5a6837e411ffa92d7265e5fd
3
+ size 14575
checkpoint-730/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e229b76163feaaaed94654afc091c2f7c4136846925b78769f9142263c1ef6e6
3
+ size 627
checkpoint-730/trainer_state.json ADDED
@@ -0,0 +1,2063 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.6579999923706055,
3
+ "best_model_checkpoint": "/scratch/kwamea/llama-output/checkpoint-720",
4
+ "epoch": 1.8766066838046274,
5
+ "eval_steps": 5,
6
+ "global_step": 730,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 9.93573264781491e-05,
14
+ "loss": 1.9824,
15
+ "step": 5
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "eval_loss": 1.954480528831482,
20
+ "eval_runtime": 38.0417,
21
+ "eval_samples_per_second": 2.208,
22
+ "eval_steps_per_second": 0.289,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.03,
27
+ "learning_rate": 9.87146529562982e-05,
28
+ "loss": 1.9245,
29
+ "step": 10
30
+ },
31
+ {
32
+ "epoch": 0.03,
33
+ "eval_loss": 1.8942010402679443,
34
+ "eval_runtime": 37.8934,
35
+ "eval_samples_per_second": 2.217,
36
+ "eval_steps_per_second": 0.29,
37
+ "step": 10
38
+ },
39
+ {
40
+ "epoch": 0.04,
41
+ "learning_rate": 9.80719794344473e-05,
42
+ "loss": 1.8616,
43
+ "step": 15
44
+ },
45
+ {
46
+ "epoch": 0.04,
47
+ "eval_loss": 1.8424055576324463,
48
+ "eval_runtime": 37.8727,
49
+ "eval_samples_per_second": 2.218,
50
+ "eval_steps_per_second": 0.29,
51
+ "step": 15
52
+ },
53
+ {
54
+ "epoch": 0.05,
55
+ "learning_rate": 9.742930591259641e-05,
56
+ "loss": 1.8267,
57
+ "step": 20
58
+ },
59
+ {
60
+ "epoch": 0.05,
61
+ "eval_loss": 1.8279685974121094,
62
+ "eval_runtime": 38.166,
63
+ "eval_samples_per_second": 2.201,
64
+ "eval_steps_per_second": 0.288,
65
+ "step": 20
66
+ },
67
+ {
68
+ "epoch": 0.06,
69
+ "learning_rate": 9.67866323907455e-05,
70
+ "loss": 1.835,
71
+ "step": 25
72
+ },
73
+ {
74
+ "epoch": 0.06,
75
+ "eval_loss": 1.8134318590164185,
76
+ "eval_runtime": 38.0411,
77
+ "eval_samples_per_second": 2.208,
78
+ "eval_steps_per_second": 0.289,
79
+ "step": 25
80
+ },
81
+ {
82
+ "epoch": 0.08,
83
+ "learning_rate": 9.61439588688946e-05,
84
+ "loss": 1.8236,
85
+ "step": 30
86
+ },
87
+ {
88
+ "epoch": 0.08,
89
+ "eval_loss": 1.8010995388031006,
90
+ "eval_runtime": 37.8753,
91
+ "eval_samples_per_second": 2.218,
92
+ "eval_steps_per_second": 0.29,
93
+ "step": 30
94
+ },
95
+ {
96
+ "epoch": 0.09,
97
+ "learning_rate": 9.550128534704372e-05,
98
+ "loss": 1.7186,
99
+ "step": 35
100
+ },
101
+ {
102
+ "epoch": 0.09,
103
+ "eval_loss": 1.7901933193206787,
104
+ "eval_runtime": 38.0374,
105
+ "eval_samples_per_second": 2.208,
106
+ "eval_steps_per_second": 0.289,
107
+ "step": 35
108
+ },
109
+ {
110
+ "epoch": 0.1,
111
+ "learning_rate": 9.485861182519281e-05,
112
+ "loss": 1.7959,
113
+ "step": 40
114
+ },
115
+ {
116
+ "epoch": 0.1,
117
+ "eval_loss": 1.7807564735412598,
118
+ "eval_runtime": 37.8784,
119
+ "eval_samples_per_second": 2.218,
120
+ "eval_steps_per_second": 0.29,
121
+ "step": 40
122
+ },
123
+ {
124
+ "epoch": 0.12,
125
+ "learning_rate": 9.421593830334192e-05,
126
+ "loss": 1.715,
127
+ "step": 45
128
+ },
129
+ {
130
+ "epoch": 0.12,
131
+ "eval_loss": 1.7709146738052368,
132
+ "eval_runtime": 37.8705,
133
+ "eval_samples_per_second": 2.218,
134
+ "eval_steps_per_second": 0.29,
135
+ "step": 45
136
+ },
137
+ {
138
+ "epoch": 0.13,
139
+ "learning_rate": 9.357326478149101e-05,
140
+ "loss": 1.7581,
141
+ "step": 50
142
+ },
143
+ {
144
+ "epoch": 0.13,
145
+ "eval_loss": 1.761339783668518,
146
+ "eval_runtime": 37.661,
147
+ "eval_samples_per_second": 2.23,
148
+ "eval_steps_per_second": 0.292,
149
+ "step": 50
150
+ },
151
+ {
152
+ "epoch": 0.14,
153
+ "learning_rate": 9.29305912596401e-05,
154
+ "loss": 1.7305,
155
+ "step": 55
156
+ },
157
+ {
158
+ "epoch": 0.14,
159
+ "eval_loss": 1.7389689683914185,
160
+ "eval_runtime": 38.0055,
161
+ "eval_samples_per_second": 2.21,
162
+ "eval_steps_per_second": 0.289,
163
+ "step": 55
164
+ },
165
+ {
166
+ "epoch": 0.15,
167
+ "learning_rate": 9.228791773778921e-05,
168
+ "loss": 1.7086,
169
+ "step": 60
170
+ },
171
+ {
172
+ "epoch": 0.15,
173
+ "eval_loss": 1.713218092918396,
174
+ "eval_runtime": 37.737,
175
+ "eval_samples_per_second": 2.226,
176
+ "eval_steps_per_second": 0.291,
177
+ "step": 60
178
+ },
179
+ {
180
+ "epoch": 0.17,
181
+ "learning_rate": 9.16452442159383e-05,
182
+ "loss": 1.7057,
183
+ "step": 65
184
+ },
185
+ {
186
+ "epoch": 0.17,
187
+ "eval_loss": 1.7108122110366821,
188
+ "eval_runtime": 38.0037,
189
+ "eval_samples_per_second": 2.21,
190
+ "eval_steps_per_second": 0.289,
191
+ "step": 65
192
+ },
193
+ {
194
+ "epoch": 0.18,
195
+ "learning_rate": 9.100257069408741e-05,
196
+ "loss": 1.7243,
197
+ "step": 70
198
+ },
199
+ {
200
+ "epoch": 0.18,
201
+ "eval_loss": 1.7041622400283813,
202
+ "eval_runtime": 38.1402,
203
+ "eval_samples_per_second": 2.202,
204
+ "eval_steps_per_second": 0.288,
205
+ "step": 70
206
+ },
207
+ {
208
+ "epoch": 0.19,
209
+ "learning_rate": 9.03598971722365e-05,
210
+ "loss": 1.7038,
211
+ "step": 75
212
+ },
213
+ {
214
+ "epoch": 0.19,
215
+ "eval_loss": 1.7010408639907837,
216
+ "eval_runtime": 37.7193,
217
+ "eval_samples_per_second": 2.227,
218
+ "eval_steps_per_second": 0.292,
219
+ "step": 75
220
+ },
221
+ {
222
+ "epoch": 0.21,
223
+ "learning_rate": 8.97172236503856e-05,
224
+ "loss": 1.6939,
225
+ "step": 80
226
+ },
227
+ {
228
+ "epoch": 0.21,
229
+ "eval_loss": 1.6987501382827759,
230
+ "eval_runtime": 38.0624,
231
+ "eval_samples_per_second": 2.207,
232
+ "eval_steps_per_second": 0.289,
233
+ "step": 80
234
+ },
235
+ {
236
+ "epoch": 0.22,
237
+ "learning_rate": 8.907455012853471e-05,
238
+ "loss": 1.645,
239
+ "step": 85
240
+ },
241
+ {
242
+ "epoch": 0.22,
243
+ "eval_loss": 1.6968218088150024,
244
+ "eval_runtime": 37.9205,
245
+ "eval_samples_per_second": 2.215,
246
+ "eval_steps_per_second": 0.29,
247
+ "step": 85
248
+ },
249
+ {
250
+ "epoch": 0.23,
251
+ "learning_rate": 8.84318766066838e-05,
252
+ "loss": 1.7016,
253
+ "step": 90
254
+ },
255
+ {
256
+ "epoch": 0.23,
257
+ "eval_loss": 1.6952500343322754,
258
+ "eval_runtime": 37.9617,
259
+ "eval_samples_per_second": 2.213,
260
+ "eval_steps_per_second": 0.29,
261
+ "step": 90
262
+ },
263
+ {
264
+ "epoch": 0.24,
265
+ "learning_rate": 8.778920308483291e-05,
266
+ "loss": 1.6869,
267
+ "step": 95
268
+ },
269
+ {
270
+ "epoch": 0.24,
271
+ "eval_loss": 1.6932854652404785,
272
+ "eval_runtime": 37.8781,
273
+ "eval_samples_per_second": 2.218,
274
+ "eval_steps_per_second": 0.29,
275
+ "step": 95
276
+ },
277
+ {
278
+ "epoch": 0.26,
279
+ "learning_rate": 8.7146529562982e-05,
280
+ "loss": 1.6728,
281
+ "step": 100
282
+ },
283
+ {
284
+ "epoch": 0.26,
285
+ "eval_loss": 1.6920738220214844,
286
+ "eval_runtime": 38.1996,
287
+ "eval_samples_per_second": 2.199,
288
+ "eval_steps_per_second": 0.288,
289
+ "step": 100
290
+ },
291
+ {
292
+ "epoch": 0.27,
293
+ "learning_rate": 8.650385604113111e-05,
294
+ "loss": 1.6674,
295
+ "step": 105
296
+ },
297
+ {
298
+ "epoch": 0.27,
299
+ "eval_loss": 1.6904469728469849,
300
+ "eval_runtime": 38.1252,
301
+ "eval_samples_per_second": 2.203,
302
+ "eval_steps_per_second": 0.289,
303
+ "step": 105
304
+ },
305
+ {
306
+ "epoch": 0.28,
307
+ "learning_rate": 8.586118251928022e-05,
308
+ "loss": 1.7013,
309
+ "step": 110
310
+ },
311
+ {
312
+ "epoch": 0.28,
313
+ "eval_loss": 1.6895124912261963,
314
+ "eval_runtime": 38.0968,
315
+ "eval_samples_per_second": 2.205,
316
+ "eval_steps_per_second": 0.289,
317
+ "step": 110
318
+ },
319
+ {
320
+ "epoch": 0.3,
321
+ "learning_rate": 8.521850899742931e-05,
322
+ "loss": 1.7148,
323
+ "step": 115
324
+ },
325
+ {
326
+ "epoch": 0.3,
327
+ "eval_loss": 1.6886374950408936,
328
+ "eval_runtime": 37.8799,
329
+ "eval_samples_per_second": 2.218,
330
+ "eval_steps_per_second": 0.29,
331
+ "step": 115
332
+ },
333
+ {
334
+ "epoch": 0.31,
335
+ "learning_rate": 8.457583547557842e-05,
336
+ "loss": 1.7166,
337
+ "step": 120
338
+ },
339
+ {
340
+ "epoch": 0.31,
341
+ "eval_loss": 1.6868164539337158,
342
+ "eval_runtime": 37.8861,
343
+ "eval_samples_per_second": 2.217,
344
+ "eval_steps_per_second": 0.29,
345
+ "step": 120
346
+ },
347
+ {
348
+ "epoch": 0.32,
349
+ "learning_rate": 8.393316195372751e-05,
350
+ "loss": 1.7012,
351
+ "step": 125
352
+ },
353
+ {
354
+ "epoch": 0.32,
355
+ "eval_loss": 1.6858367919921875,
356
+ "eval_runtime": 38.1136,
357
+ "eval_samples_per_second": 2.204,
358
+ "eval_steps_per_second": 0.289,
359
+ "step": 125
360
+ },
361
+ {
362
+ "epoch": 0.33,
363
+ "learning_rate": 8.32904884318766e-05,
364
+ "loss": 1.6827,
365
+ "step": 130
366
+ },
367
+ {
368
+ "epoch": 0.33,
369
+ "eval_loss": 1.6849361658096313,
370
+ "eval_runtime": 37.9883,
371
+ "eval_samples_per_second": 2.211,
372
+ "eval_steps_per_second": 0.29,
373
+ "step": 130
374
+ },
375
+ {
376
+ "epoch": 0.35,
377
+ "learning_rate": 8.264781491002571e-05,
378
+ "loss": 1.7157,
379
+ "step": 135
380
+ },
381
+ {
382
+ "epoch": 0.35,
383
+ "eval_loss": 1.684756875038147,
384
+ "eval_runtime": 37.8726,
385
+ "eval_samples_per_second": 2.218,
386
+ "eval_steps_per_second": 0.29,
387
+ "step": 135
388
+ },
389
+ {
390
+ "epoch": 0.36,
391
+ "learning_rate": 8.200514138817481e-05,
392
+ "loss": 1.6668,
393
+ "step": 140
394
+ },
395
+ {
396
+ "epoch": 0.36,
397
+ "eval_loss": 1.6836236715316772,
398
+ "eval_runtime": 37.9586,
399
+ "eval_samples_per_second": 2.213,
400
+ "eval_steps_per_second": 0.29,
401
+ "step": 140
402
+ },
403
+ {
404
+ "epoch": 0.37,
405
+ "learning_rate": 8.136246786632391e-05,
406
+ "loss": 1.6588,
407
+ "step": 145
408
+ },
409
+ {
410
+ "epoch": 0.37,
411
+ "eval_loss": 1.6828166246414185,
412
+ "eval_runtime": 38.0596,
413
+ "eval_samples_per_second": 2.207,
414
+ "eval_steps_per_second": 0.289,
415
+ "step": 145
416
+ },
417
+ {
418
+ "epoch": 0.39,
419
+ "learning_rate": 8.071979434447301e-05,
420
+ "loss": 1.7005,
421
+ "step": 150
422
+ },
423
+ {
424
+ "epoch": 0.39,
425
+ "eval_loss": 1.681463360786438,
426
+ "eval_runtime": 38.0878,
427
+ "eval_samples_per_second": 2.205,
428
+ "eval_steps_per_second": 0.289,
429
+ "step": 150
430
+ },
431
+ {
432
+ "epoch": 0.4,
433
+ "learning_rate": 8.007712082262212e-05,
434
+ "loss": 1.6893,
435
+ "step": 155
436
+ },
437
+ {
438
+ "epoch": 0.4,
439
+ "eval_loss": 1.6811552047729492,
440
+ "eval_runtime": 37.9265,
441
+ "eval_samples_per_second": 2.215,
442
+ "eval_steps_per_second": 0.29,
443
+ "step": 155
444
+ },
445
+ {
446
+ "epoch": 0.41,
447
+ "learning_rate": 7.943444730077121e-05,
448
+ "loss": 1.6682,
449
+ "step": 160
450
+ },
451
+ {
452
+ "epoch": 0.41,
453
+ "eval_loss": 1.6801096200942993,
454
+ "eval_runtime": 38.0375,
455
+ "eval_samples_per_second": 2.208,
456
+ "eval_steps_per_second": 0.289,
457
+ "step": 160
458
+ },
459
+ {
460
+ "epoch": 0.42,
461
+ "learning_rate": 7.87917737789203e-05,
462
+ "loss": 1.6481,
463
+ "step": 165
464
+ },
465
+ {
466
+ "epoch": 0.42,
467
+ "eval_loss": 1.6800144910812378,
468
+ "eval_runtime": 37.8916,
469
+ "eval_samples_per_second": 2.217,
470
+ "eval_steps_per_second": 0.29,
471
+ "step": 165
472
+ },
473
+ {
474
+ "epoch": 0.44,
475
+ "learning_rate": 7.814910025706941e-05,
476
+ "loss": 1.7042,
477
+ "step": 170
478
+ },
479
+ {
480
+ "epoch": 0.44,
481
+ "eval_loss": 1.6784976720809937,
482
+ "eval_runtime": 37.8514,
483
+ "eval_samples_per_second": 2.219,
484
+ "eval_steps_per_second": 0.291,
485
+ "step": 170
486
+ },
487
+ {
488
+ "epoch": 0.45,
489
+ "learning_rate": 7.750642673521852e-05,
490
+ "loss": 1.6555,
491
+ "step": 175
492
+ },
493
+ {
494
+ "epoch": 0.45,
495
+ "eval_loss": 1.6780468225479126,
496
+ "eval_runtime": 37.8893,
497
+ "eval_samples_per_second": 2.217,
498
+ "eval_steps_per_second": 0.29,
499
+ "step": 175
500
+ },
501
+ {
502
+ "epoch": 0.46,
503
+ "learning_rate": 7.686375321336761e-05,
504
+ "loss": 1.6717,
505
+ "step": 180
506
+ },
507
+ {
508
+ "epoch": 0.46,
509
+ "eval_loss": 1.6775085926055908,
510
+ "eval_runtime": 37.9289,
511
+ "eval_samples_per_second": 2.215,
512
+ "eval_steps_per_second": 0.29,
513
+ "step": 180
514
+ },
515
+ {
516
+ "epoch": 0.48,
517
+ "learning_rate": 7.622107969151672e-05,
518
+ "loss": 1.6716,
519
+ "step": 185
520
+ },
521
+ {
522
+ "epoch": 0.48,
523
+ "eval_loss": 1.6770914793014526,
524
+ "eval_runtime": 38.0111,
525
+ "eval_samples_per_second": 2.21,
526
+ "eval_steps_per_second": 0.289,
527
+ "step": 185
528
+ },
529
+ {
530
+ "epoch": 0.49,
531
+ "learning_rate": 7.557840616966581e-05,
532
+ "loss": 1.6618,
533
+ "step": 190
534
+ },
535
+ {
536
+ "epoch": 0.49,
537
+ "eval_loss": 1.676563024520874,
538
+ "eval_runtime": 38.0703,
539
+ "eval_samples_per_second": 2.206,
540
+ "eval_steps_per_second": 0.289,
541
+ "step": 190
542
+ },
543
+ {
544
+ "epoch": 0.5,
545
+ "learning_rate": 7.493573264781492e-05,
546
+ "loss": 1.6411,
547
+ "step": 195
548
+ },
549
+ {
550
+ "epoch": 0.5,
551
+ "eval_loss": 1.6757832765579224,
552
+ "eval_runtime": 37.827,
553
+ "eval_samples_per_second": 2.221,
554
+ "eval_steps_per_second": 0.291,
555
+ "step": 195
556
+ },
557
+ {
558
+ "epoch": 0.51,
559
+ "learning_rate": 7.429305912596401e-05,
560
+ "loss": 1.6637,
561
+ "step": 200
562
+ },
563
+ {
564
+ "epoch": 0.51,
565
+ "eval_loss": 1.6751985549926758,
566
+ "eval_runtime": 37.836,
567
+ "eval_samples_per_second": 2.22,
568
+ "eval_steps_per_second": 0.291,
569
+ "step": 200
570
+ },
571
+ {
572
+ "epoch": 0.53,
573
+ "learning_rate": 7.365038560411311e-05,
574
+ "loss": 1.6142,
575
+ "step": 205
576
+ },
577
+ {
578
+ "epoch": 0.53,
579
+ "eval_loss": 1.675147294998169,
580
+ "eval_runtime": 37.9175,
581
+ "eval_samples_per_second": 2.215,
582
+ "eval_steps_per_second": 0.29,
583
+ "step": 205
584
+ },
585
+ {
586
+ "epoch": 0.54,
587
+ "learning_rate": 7.300771208226222e-05,
588
+ "loss": 1.6741,
589
+ "step": 210
590
+ },
591
+ {
592
+ "epoch": 0.54,
593
+ "eval_loss": 1.674437165260315,
594
+ "eval_runtime": 38.087,
595
+ "eval_samples_per_second": 2.205,
596
+ "eval_steps_per_second": 0.289,
597
+ "step": 210
598
+ },
599
+ {
600
+ "epoch": 0.55,
601
+ "learning_rate": 7.236503856041131e-05,
602
+ "loss": 1.6408,
603
+ "step": 215
604
+ },
605
+ {
606
+ "epoch": 0.55,
607
+ "eval_loss": 1.6737719774246216,
608
+ "eval_runtime": 38.0075,
609
+ "eval_samples_per_second": 2.21,
610
+ "eval_steps_per_second": 0.289,
611
+ "step": 215
612
+ },
613
+ {
614
+ "epoch": 0.57,
615
+ "learning_rate": 7.172236503856042e-05,
616
+ "loss": 1.6733,
617
+ "step": 220
618
+ },
619
+ {
620
+ "epoch": 0.57,
621
+ "eval_loss": 1.6732759475708008,
622
+ "eval_runtime": 38.1125,
623
+ "eval_samples_per_second": 2.204,
624
+ "eval_steps_per_second": 0.289,
625
+ "step": 220
626
+ },
627
+ {
628
+ "epoch": 0.58,
629
+ "learning_rate": 7.107969151670951e-05,
630
+ "loss": 1.679,
631
+ "step": 225
632
+ },
633
+ {
634
+ "epoch": 0.58,
635
+ "eval_loss": 1.6726195812225342,
636
+ "eval_runtime": 37.9769,
637
+ "eval_samples_per_second": 2.212,
638
+ "eval_steps_per_second": 0.29,
639
+ "step": 225
640
+ },
641
+ {
642
+ "epoch": 0.59,
643
+ "learning_rate": 7.043701799485862e-05,
644
+ "loss": 1.7202,
645
+ "step": 230
646
+ },
647
+ {
648
+ "epoch": 0.59,
649
+ "eval_loss": 1.671908974647522,
650
+ "eval_runtime": 37.8215,
651
+ "eval_samples_per_second": 2.221,
652
+ "eval_steps_per_second": 0.291,
653
+ "step": 230
654
+ },
655
+ {
656
+ "epoch": 0.6,
657
+ "learning_rate": 6.979434447300771e-05,
658
+ "loss": 1.6805,
659
+ "step": 235
660
+ },
661
+ {
662
+ "epoch": 0.6,
663
+ "eval_loss": 1.6715577840805054,
664
+ "eval_runtime": 37.9343,
665
+ "eval_samples_per_second": 2.214,
666
+ "eval_steps_per_second": 0.29,
667
+ "step": 235
668
+ },
669
+ {
670
+ "epoch": 0.62,
671
+ "learning_rate": 6.91516709511568e-05,
672
+ "loss": 1.6331,
673
+ "step": 240
674
+ },
675
+ {
676
+ "epoch": 0.62,
677
+ "eval_loss": 1.6715376377105713,
678
+ "eval_runtime": 37.9904,
679
+ "eval_samples_per_second": 2.211,
680
+ "eval_steps_per_second": 0.29,
681
+ "step": 240
682
+ },
683
+ {
684
+ "epoch": 0.63,
685
+ "learning_rate": 6.850899742930593e-05,
686
+ "loss": 1.6761,
687
+ "step": 245
688
+ },
689
+ {
690
+ "epoch": 0.63,
691
+ "eval_loss": 1.671446681022644,
692
+ "eval_runtime": 38.0357,
693
+ "eval_samples_per_second": 2.208,
694
+ "eval_steps_per_second": 0.289,
695
+ "step": 245
696
+ },
697
+ {
698
+ "epoch": 0.64,
699
+ "learning_rate": 6.786632390745502e-05,
700
+ "loss": 1.6994,
701
+ "step": 250
702
+ },
703
+ {
704
+ "epoch": 0.64,
705
+ "eval_loss": 1.6716604232788086,
706
+ "eval_runtime": 37.9609,
707
+ "eval_samples_per_second": 2.213,
708
+ "eval_steps_per_second": 0.29,
709
+ "step": 250
710
+ },
711
+ {
712
+ "epoch": 0.66,
713
+ "learning_rate": 6.722365038560411e-05,
714
+ "loss": 1.6305,
715
+ "step": 255
716
+ },
717
+ {
718
+ "epoch": 0.66,
719
+ "eval_loss": 1.6711723804473877,
720
+ "eval_runtime": 37.7893,
721
+ "eval_samples_per_second": 2.223,
722
+ "eval_steps_per_second": 0.291,
723
+ "step": 255
724
+ },
725
+ {
726
+ "epoch": 0.67,
727
+ "learning_rate": 6.658097686375322e-05,
728
+ "loss": 1.6612,
729
+ "step": 260
730
+ },
731
+ {
732
+ "epoch": 0.67,
733
+ "eval_loss": 1.670398473739624,
734
+ "eval_runtime": 37.7667,
735
+ "eval_samples_per_second": 2.224,
736
+ "eval_steps_per_second": 0.291,
737
+ "step": 260
738
+ },
739
+ {
740
+ "epoch": 0.68,
741
+ "learning_rate": 6.593830334190231e-05,
742
+ "loss": 1.6576,
743
+ "step": 265
744
+ },
745
+ {
746
+ "epoch": 0.68,
747
+ "eval_loss": 1.6706459522247314,
748
+ "eval_runtime": 37.8285,
749
+ "eval_samples_per_second": 2.221,
750
+ "eval_steps_per_second": 0.291,
751
+ "step": 265
752
+ },
753
+ {
754
+ "epoch": 0.69,
755
+ "learning_rate": 6.529562982005142e-05,
756
+ "loss": 1.6837,
757
+ "step": 270
758
+ },
759
+ {
760
+ "epoch": 0.69,
761
+ "eval_loss": 1.6699285507202148,
762
+ "eval_runtime": 37.7619,
763
+ "eval_samples_per_second": 2.224,
764
+ "eval_steps_per_second": 0.291,
765
+ "step": 270
766
+ },
767
+ {
768
+ "epoch": 0.71,
769
+ "learning_rate": 6.465295629820052e-05,
770
+ "loss": 1.6493,
771
+ "step": 275
772
+ },
773
+ {
774
+ "epoch": 0.71,
775
+ "eval_loss": 1.6693135499954224,
776
+ "eval_runtime": 37.7441,
777
+ "eval_samples_per_second": 2.226,
778
+ "eval_steps_per_second": 0.291,
779
+ "step": 275
780
+ },
781
+ {
782
+ "epoch": 0.72,
783
+ "learning_rate": 6.401028277634962e-05,
784
+ "loss": 1.6427,
785
+ "step": 280
786
+ },
787
+ {
788
+ "epoch": 0.72,
789
+ "eval_loss": 1.669467568397522,
790
+ "eval_runtime": 37.8707,
791
+ "eval_samples_per_second": 2.218,
792
+ "eval_steps_per_second": 0.29,
793
+ "step": 280
794
+ },
795
+ {
796
+ "epoch": 0.73,
797
+ "learning_rate": 6.336760925449872e-05,
798
+ "loss": 1.6236,
799
+ "step": 285
800
+ },
801
+ {
802
+ "epoch": 0.73,
803
+ "eval_loss": 1.6690431833267212,
804
+ "eval_runtime": 37.8946,
805
+ "eval_samples_per_second": 2.217,
806
+ "eval_steps_per_second": 0.29,
807
+ "step": 285
808
+ },
809
+ {
810
+ "epoch": 0.75,
811
+ "learning_rate": 6.272493573264781e-05,
812
+ "loss": 1.6538,
813
+ "step": 290
814
+ },
815
+ {
816
+ "epoch": 0.75,
817
+ "eval_loss": 1.6688117980957031,
818
+ "eval_runtime": 37.9378,
819
+ "eval_samples_per_second": 2.214,
820
+ "eval_steps_per_second": 0.29,
821
+ "step": 290
822
+ },
823
+ {
824
+ "epoch": 0.76,
825
+ "learning_rate": 6.208226221079692e-05,
826
+ "loss": 1.6558,
827
+ "step": 295
828
+ },
829
+ {
830
+ "epoch": 0.76,
831
+ "eval_loss": 1.668560266494751,
832
+ "eval_runtime": 38.0448,
833
+ "eval_samples_per_second": 2.208,
834
+ "eval_steps_per_second": 0.289,
835
+ "step": 295
836
+ },
837
+ {
838
+ "epoch": 0.77,
839
+ "learning_rate": 6.143958868894601e-05,
840
+ "loss": 1.6401,
841
+ "step": 300
842
+ },
843
+ {
844
+ "epoch": 0.77,
845
+ "eval_loss": 1.6680612564086914,
846
+ "eval_runtime": 37.8571,
847
+ "eval_samples_per_second": 2.219,
848
+ "eval_steps_per_second": 0.291,
849
+ "step": 300
850
+ },
851
+ {
852
+ "epoch": 0.78,
853
+ "learning_rate": 6.079691516709511e-05,
854
+ "loss": 1.6468,
855
+ "step": 305
856
+ },
857
+ {
858
+ "epoch": 0.78,
859
+ "eval_loss": 1.6675914525985718,
860
+ "eval_runtime": 37.8491,
861
+ "eval_samples_per_second": 2.219,
862
+ "eval_steps_per_second": 0.291,
863
+ "step": 305
864
+ },
865
+ {
866
+ "epoch": 0.8,
867
+ "learning_rate": 6.015424164524421e-05,
868
+ "loss": 1.6579,
869
+ "step": 310
870
+ },
871
+ {
872
+ "epoch": 0.8,
873
+ "eval_loss": 1.6671632528305054,
874
+ "eval_runtime": 37.7658,
875
+ "eval_samples_per_second": 2.224,
876
+ "eval_steps_per_second": 0.291,
877
+ "step": 310
878
+ },
879
+ {
880
+ "epoch": 0.81,
881
+ "learning_rate": 5.951156812339333e-05,
882
+ "loss": 1.6339,
883
+ "step": 315
884
+ },
885
+ {
886
+ "epoch": 0.81,
887
+ "eval_loss": 1.666803240776062,
888
+ "eval_runtime": 37.7896,
889
+ "eval_samples_per_second": 2.223,
890
+ "eval_steps_per_second": 0.291,
891
+ "step": 315
892
+ },
893
+ {
894
+ "epoch": 0.82,
895
+ "learning_rate": 5.886889460154242e-05,
896
+ "loss": 1.6636,
897
+ "step": 320
898
+ },
899
+ {
900
+ "epoch": 0.82,
901
+ "eval_loss": 1.6664105653762817,
902
+ "eval_runtime": 38.0586,
903
+ "eval_samples_per_second": 2.207,
904
+ "eval_steps_per_second": 0.289,
905
+ "step": 320
906
+ },
907
+ {
908
+ "epoch": 0.84,
909
+ "learning_rate": 5.822622107969152e-05,
910
+ "loss": 1.641,
911
+ "step": 325
912
+ },
913
+ {
914
+ "epoch": 0.84,
915
+ "eval_loss": 1.6665875911712646,
916
+ "eval_runtime": 37.9046,
917
+ "eval_samples_per_second": 2.216,
918
+ "eval_steps_per_second": 0.29,
919
+ "step": 325
920
+ },
921
+ {
922
+ "epoch": 0.85,
923
+ "learning_rate": 5.758354755784062e-05,
924
+ "loss": 1.6616,
925
+ "step": 330
926
+ },
927
+ {
928
+ "epoch": 0.85,
929
+ "eval_loss": 1.6664165258407593,
930
+ "eval_runtime": 37.8575,
931
+ "eval_samples_per_second": 2.219,
932
+ "eval_steps_per_second": 0.291,
933
+ "step": 330
934
+ },
935
+ {
936
+ "epoch": 0.86,
937
+ "learning_rate": 5.694087403598972e-05,
938
+ "loss": 1.6597,
939
+ "step": 335
940
+ },
941
+ {
942
+ "epoch": 0.86,
943
+ "eval_loss": 1.666045904159546,
944
+ "eval_runtime": 37.9711,
945
+ "eval_samples_per_second": 2.212,
946
+ "eval_steps_per_second": 0.29,
947
+ "step": 335
948
+ },
949
+ {
950
+ "epoch": 0.87,
951
+ "learning_rate": 5.6298200514138824e-05,
952
+ "loss": 1.67,
953
+ "step": 340
954
+ },
955
+ {
956
+ "epoch": 0.87,
957
+ "eval_loss": 1.666337490081787,
958
+ "eval_runtime": 37.8821,
959
+ "eval_samples_per_second": 2.217,
960
+ "eval_steps_per_second": 0.29,
961
+ "step": 340
962
+ },
963
+ {
964
+ "epoch": 0.89,
965
+ "learning_rate": 5.5655526992287924e-05,
966
+ "loss": 1.6344,
967
+ "step": 345
968
+ },
969
+ {
970
+ "epoch": 0.89,
971
+ "eval_loss": 1.6659029722213745,
972
+ "eval_runtime": 37.9188,
973
+ "eval_samples_per_second": 2.215,
974
+ "eval_steps_per_second": 0.29,
975
+ "step": 345
976
+ },
977
+ {
978
+ "epoch": 0.9,
979
+ "learning_rate": 5.501285347043702e-05,
980
+ "loss": 1.6623,
981
+ "step": 350
982
+ },
983
+ {
984
+ "epoch": 0.9,
985
+ "eval_loss": 1.6656525135040283,
986
+ "eval_runtime": 37.9951,
987
+ "eval_samples_per_second": 2.211,
988
+ "eval_steps_per_second": 0.29,
989
+ "step": 350
990
+ },
991
+ {
992
+ "epoch": 0.91,
993
+ "learning_rate": 5.437017994858612e-05,
994
+ "loss": 1.6623,
995
+ "step": 355
996
+ },
997
+ {
998
+ "epoch": 0.91,
999
+ "eval_loss": 1.6654421091079712,
1000
+ "eval_runtime": 37.8803,
1001
+ "eval_samples_per_second": 2.218,
1002
+ "eval_steps_per_second": 0.29,
1003
+ "step": 355
1004
+ },
1005
+ {
1006
+ "epoch": 0.93,
1007
+ "learning_rate": 5.372750642673522e-05,
1008
+ "loss": 1.6741,
1009
+ "step": 360
1010
+ },
1011
+ {
1012
+ "epoch": 0.93,
1013
+ "eval_loss": 1.665541648864746,
1014
+ "eval_runtime": 37.7151,
1015
+ "eval_samples_per_second": 2.227,
1016
+ "eval_steps_per_second": 0.292,
1017
+ "step": 360
1018
+ },
1019
+ {
1020
+ "epoch": 0.94,
1021
+ "learning_rate": 5.308483290488432e-05,
1022
+ "loss": 1.6973,
1023
+ "step": 365
1024
+ },
1025
+ {
1026
+ "epoch": 0.94,
1027
+ "eval_loss": 1.6651471853256226,
1028
+ "eval_runtime": 37.9144,
1029
+ "eval_samples_per_second": 2.216,
1030
+ "eval_steps_per_second": 0.29,
1031
+ "step": 365
1032
+ },
1033
+ {
1034
+ "epoch": 0.95,
1035
+ "learning_rate": 5.244215938303342e-05,
1036
+ "loss": 1.6829,
1037
+ "step": 370
1038
+ },
1039
+ {
1040
+ "epoch": 0.95,
1041
+ "eval_loss": 1.6651064157485962,
1042
+ "eval_runtime": 37.7443,
1043
+ "eval_samples_per_second": 2.226,
1044
+ "eval_steps_per_second": 0.291,
1045
+ "step": 370
1046
+ },
1047
+ {
1048
+ "epoch": 0.96,
1049
+ "learning_rate": 5.1799485861182514e-05,
1050
+ "loss": 1.6923,
1051
+ "step": 375
1052
+ },
1053
+ {
1054
+ "epoch": 0.96,
1055
+ "eval_loss": 1.6646034717559814,
1056
+ "eval_runtime": 37.8783,
1057
+ "eval_samples_per_second": 2.218,
1058
+ "eval_steps_per_second": 0.29,
1059
+ "step": 375
1060
+ },
1061
+ {
1062
+ "epoch": 0.98,
1063
+ "learning_rate": 5.1156812339331615e-05,
1064
+ "loss": 1.6725,
1065
+ "step": 380
1066
+ },
1067
+ {
1068
+ "epoch": 0.98,
1069
+ "eval_loss": 1.6639235019683838,
1070
+ "eval_runtime": 37.7559,
1071
+ "eval_samples_per_second": 2.225,
1072
+ "eval_steps_per_second": 0.291,
1073
+ "step": 380
1074
+ },
1075
+ {
1076
+ "epoch": 0.99,
1077
+ "learning_rate": 5.051413881748073e-05,
1078
+ "loss": 1.6216,
1079
+ "step": 385
1080
+ },
1081
+ {
1082
+ "epoch": 0.99,
1083
+ "eval_loss": 1.6643506288528442,
1084
+ "eval_runtime": 37.9895,
1085
+ "eval_samples_per_second": 2.211,
1086
+ "eval_steps_per_second": 0.29,
1087
+ "step": 385
1088
+ },
1089
+ {
1090
+ "epoch": 1.0,
1091
+ "learning_rate": 4.987146529562982e-05,
1092
+ "loss": 1.6518,
1093
+ "step": 390
1094
+ },
1095
+ {
1096
+ "epoch": 1.0,
1097
+ "eval_loss": 1.6643046140670776,
1098
+ "eval_runtime": 37.9289,
1099
+ "eval_samples_per_second": 2.215,
1100
+ "eval_steps_per_second": 0.29,
1101
+ "step": 390
1102
+ },
1103
+ {
1104
+ "epoch": 1.02,
1105
+ "learning_rate": 4.922879177377892e-05,
1106
+ "loss": 1.658,
1107
+ "step": 395
1108
+ },
1109
+ {
1110
+ "epoch": 1.02,
1111
+ "eval_loss": 1.6641957759857178,
1112
+ "eval_runtime": 37.8638,
1113
+ "eval_samples_per_second": 2.218,
1114
+ "eval_steps_per_second": 0.291,
1115
+ "step": 395
1116
+ },
1117
+ {
1118
+ "epoch": 1.03,
1119
+ "learning_rate": 4.8586118251928024e-05,
1120
+ "loss": 1.6767,
1121
+ "step": 400
1122
+ },
1123
+ {
1124
+ "epoch": 1.03,
1125
+ "eval_loss": 1.663757085800171,
1126
+ "eval_runtime": 37.7728,
1127
+ "eval_samples_per_second": 2.224,
1128
+ "eval_steps_per_second": 0.291,
1129
+ "step": 400
1130
+ },
1131
+ {
1132
+ "epoch": 1.04,
1133
+ "learning_rate": 4.7943444730077124e-05,
1134
+ "loss": 1.6264,
1135
+ "step": 405
1136
+ },
1137
+ {
1138
+ "epoch": 1.04,
1139
+ "eval_loss": 1.6635041236877441,
1140
+ "eval_runtime": 37.7702,
1141
+ "eval_samples_per_second": 2.224,
1142
+ "eval_steps_per_second": 0.291,
1143
+ "step": 405
1144
+ },
1145
+ {
1146
+ "epoch": 1.05,
1147
+ "learning_rate": 4.7300771208226225e-05,
1148
+ "loss": 1.6527,
1149
+ "step": 410
1150
+ },
1151
+ {
1152
+ "epoch": 1.05,
1153
+ "eval_loss": 1.6632280349731445,
1154
+ "eval_runtime": 37.8453,
1155
+ "eval_samples_per_second": 2.22,
1156
+ "eval_steps_per_second": 0.291,
1157
+ "step": 410
1158
+ },
1159
+ {
1160
+ "epoch": 1.07,
1161
+ "learning_rate": 4.6658097686375325e-05,
1162
+ "loss": 1.6157,
1163
+ "step": 415
1164
+ },
1165
+ {
1166
+ "epoch": 1.07,
1167
+ "eval_loss": 1.6636598110198975,
1168
+ "eval_runtime": 37.812,
1169
+ "eval_samples_per_second": 2.222,
1170
+ "eval_steps_per_second": 0.291,
1171
+ "step": 415
1172
+ },
1173
+ {
1174
+ "epoch": 1.08,
1175
+ "learning_rate": 4.6015424164524426e-05,
1176
+ "loss": 1.5966,
1177
+ "step": 420
1178
+ },
1179
+ {
1180
+ "epoch": 1.08,
1181
+ "eval_loss": 1.663393497467041,
1182
+ "eval_runtime": 37.7743,
1183
+ "eval_samples_per_second": 2.224,
1184
+ "eval_steps_per_second": 0.291,
1185
+ "step": 420
1186
+ },
1187
+ {
1188
+ "epoch": 1.09,
1189
+ "learning_rate": 4.537275064267352e-05,
1190
+ "loss": 1.6705,
1191
+ "step": 425
1192
+ },
1193
+ {
1194
+ "epoch": 1.09,
1195
+ "eval_loss": 1.6631269454956055,
1196
+ "eval_runtime": 38.0953,
1197
+ "eval_samples_per_second": 2.205,
1198
+ "eval_steps_per_second": 0.289,
1199
+ "step": 425
1200
+ },
1201
+ {
1202
+ "epoch": 1.11,
1203
+ "learning_rate": 4.473007712082262e-05,
1204
+ "loss": 1.6691,
1205
+ "step": 430
1206
+ },
1207
+ {
1208
+ "epoch": 1.11,
1209
+ "eval_loss": 1.6633110046386719,
1210
+ "eval_runtime": 37.8964,
1211
+ "eval_samples_per_second": 2.217,
1212
+ "eval_steps_per_second": 0.29,
1213
+ "step": 430
1214
+ },
1215
+ {
1216
+ "epoch": 1.12,
1217
+ "learning_rate": 4.408740359897173e-05,
1218
+ "loss": 1.6332,
1219
+ "step": 435
1220
+ },
1221
+ {
1222
+ "epoch": 1.12,
1223
+ "eval_loss": 1.6628649234771729,
1224
+ "eval_runtime": 38.0621,
1225
+ "eval_samples_per_second": 2.207,
1226
+ "eval_steps_per_second": 0.289,
1227
+ "step": 435
1228
+ },
1229
+ {
1230
+ "epoch": 1.13,
1231
+ "learning_rate": 4.344473007712083e-05,
1232
+ "loss": 1.5916,
1233
+ "step": 440
1234
+ },
1235
+ {
1236
+ "epoch": 1.13,
1237
+ "eval_loss": 1.662834882736206,
1238
+ "eval_runtime": 37.9722,
1239
+ "eval_samples_per_second": 2.212,
1240
+ "eval_steps_per_second": 0.29,
1241
+ "step": 440
1242
+ },
1243
+ {
1244
+ "epoch": 1.14,
1245
+ "learning_rate": 4.280205655526993e-05,
1246
+ "loss": 1.6543,
1247
+ "step": 445
1248
+ },
1249
+ {
1250
+ "epoch": 1.14,
1251
+ "eval_loss": 1.6630265712738037,
1252
+ "eval_runtime": 37.9699,
1253
+ "eval_samples_per_second": 2.212,
1254
+ "eval_steps_per_second": 0.29,
1255
+ "step": 445
1256
+ },
1257
+ {
1258
+ "epoch": 1.16,
1259
+ "learning_rate": 4.215938303341902e-05,
1260
+ "loss": 1.6353,
1261
+ "step": 450
1262
+ },
1263
+ {
1264
+ "epoch": 1.16,
1265
+ "eval_loss": 1.6625033617019653,
1266
+ "eval_runtime": 37.7505,
1267
+ "eval_samples_per_second": 2.225,
1268
+ "eval_steps_per_second": 0.291,
1269
+ "step": 450
1270
+ },
1271
+ {
1272
+ "epoch": 1.17,
1273
+ "learning_rate": 4.151670951156812e-05,
1274
+ "loss": 1.6441,
1275
+ "step": 455
1276
+ },
1277
+ {
1278
+ "epoch": 1.17,
1279
+ "eval_loss": 1.662785530090332,
1280
+ "eval_runtime": 37.75,
1281
+ "eval_samples_per_second": 2.225,
1282
+ "eval_steps_per_second": 0.291,
1283
+ "step": 455
1284
+ },
1285
+ {
1286
+ "epoch": 1.18,
1287
+ "learning_rate": 4.0874035989717224e-05,
1288
+ "loss": 1.6631,
1289
+ "step": 460
1290
+ },
1291
+ {
1292
+ "epoch": 1.18,
1293
+ "eval_loss": 1.6627779006958008,
1294
+ "eval_runtime": 37.864,
1295
+ "eval_samples_per_second": 2.218,
1296
+ "eval_steps_per_second": 0.291,
1297
+ "step": 460
1298
+ },
1299
+ {
1300
+ "epoch": 1.2,
1301
+ "learning_rate": 4.0231362467866324e-05,
1302
+ "loss": 1.6327,
1303
+ "step": 465
1304
+ },
1305
+ {
1306
+ "epoch": 1.2,
1307
+ "eval_loss": 1.662458896636963,
1308
+ "eval_runtime": 37.7496,
1309
+ "eval_samples_per_second": 2.225,
1310
+ "eval_steps_per_second": 0.291,
1311
+ "step": 465
1312
+ },
1313
+ {
1314
+ "epoch": 1.21,
1315
+ "learning_rate": 3.958868894601543e-05,
1316
+ "loss": 1.6238,
1317
+ "step": 470
1318
+ },
1319
+ {
1320
+ "epoch": 1.21,
1321
+ "eval_loss": 1.6621876955032349,
1322
+ "eval_runtime": 37.8943,
1323
+ "eval_samples_per_second": 2.217,
1324
+ "eval_steps_per_second": 0.29,
1325
+ "step": 470
1326
+ },
1327
+ {
1328
+ "epoch": 1.22,
1329
+ "learning_rate": 3.8946015424164526e-05,
1330
+ "loss": 1.6231,
1331
+ "step": 475
1332
+ },
1333
+ {
1334
+ "epoch": 1.22,
1335
+ "eval_loss": 1.662013292312622,
1336
+ "eval_runtime": 37.9364,
1337
+ "eval_samples_per_second": 2.214,
1338
+ "eval_steps_per_second": 0.29,
1339
+ "step": 475
1340
+ },
1341
+ {
1342
+ "epoch": 1.23,
1343
+ "learning_rate": 3.8303341902313626e-05,
1344
+ "loss": 1.6381,
1345
+ "step": 480
1346
+ },
1347
+ {
1348
+ "epoch": 1.23,
1349
+ "eval_loss": 1.6615785360336304,
1350
+ "eval_runtime": 37.91,
1351
+ "eval_samples_per_second": 2.216,
1352
+ "eval_steps_per_second": 0.29,
1353
+ "step": 480
1354
+ },
1355
+ {
1356
+ "epoch": 1.25,
1357
+ "learning_rate": 3.766066838046273e-05,
1358
+ "loss": 1.6419,
1359
+ "step": 485
1360
+ },
1361
+ {
1362
+ "epoch": 1.25,
1363
+ "eval_loss": 1.662042498588562,
1364
+ "eval_runtime": 37.7834,
1365
+ "eval_samples_per_second": 2.223,
1366
+ "eval_steps_per_second": 0.291,
1367
+ "step": 485
1368
+ },
1369
+ {
1370
+ "epoch": 1.26,
1371
+ "learning_rate": 3.701799485861183e-05,
1372
+ "loss": 1.6319,
1373
+ "step": 490
1374
+ },
1375
+ {
1376
+ "epoch": 1.26,
1377
+ "eval_loss": 1.6619195938110352,
1378
+ "eval_runtime": 37.7385,
1379
+ "eval_samples_per_second": 2.226,
1380
+ "eval_steps_per_second": 0.291,
1381
+ "step": 490
1382
+ },
1383
+ {
1384
+ "epoch": 1.27,
1385
+ "learning_rate": 3.637532133676093e-05,
1386
+ "loss": 1.609,
1387
+ "step": 495
1388
+ },
1389
+ {
1390
+ "epoch": 1.27,
1391
+ "eval_loss": 1.6615225076675415,
1392
+ "eval_runtime": 37.8056,
1393
+ "eval_samples_per_second": 2.222,
1394
+ "eval_steps_per_second": 0.291,
1395
+ "step": 495
1396
+ },
1397
+ {
1398
+ "epoch": 1.29,
1399
+ "learning_rate": 3.573264781491003e-05,
1400
+ "loss": 1.6435,
1401
+ "step": 500
1402
+ },
1403
+ {
1404
+ "epoch": 1.29,
1405
+ "eval_loss": 1.6610548496246338,
1406
+ "eval_runtime": 37.9353,
1407
+ "eval_samples_per_second": 2.214,
1408
+ "eval_steps_per_second": 0.29,
1409
+ "step": 500
1410
+ },
1411
+ {
1412
+ "epoch": 1.3,
1413
+ "learning_rate": 3.508997429305913e-05,
1414
+ "loss": 1.6381,
1415
+ "step": 505
1416
+ },
1417
+ {
1418
+ "epoch": 1.3,
1419
+ "eval_loss": 1.661049246788025,
1420
+ "eval_runtime": 37.9669,
1421
+ "eval_samples_per_second": 2.212,
1422
+ "eval_steps_per_second": 0.29,
1423
+ "step": 505
1424
+ },
1425
+ {
1426
+ "epoch": 1.31,
1427
+ "learning_rate": 3.444730077120823e-05,
1428
+ "loss": 1.6506,
1429
+ "step": 510
1430
+ },
1431
+ {
1432
+ "epoch": 1.31,
1433
+ "eval_loss": 1.6610089540481567,
1434
+ "eval_runtime": 37.9127,
1435
+ "eval_samples_per_second": 2.216,
1436
+ "eval_steps_per_second": 0.29,
1437
+ "step": 510
1438
+ },
1439
+ {
1440
+ "epoch": 1.32,
1441
+ "learning_rate": 3.380462724935733e-05,
1442
+ "loss": 1.6376,
1443
+ "step": 515
1444
+ },
1445
+ {
1446
+ "epoch": 1.32,
1447
+ "eval_loss": 1.6608952283859253,
1448
+ "eval_runtime": 37.9454,
1449
+ "eval_samples_per_second": 2.214,
1450
+ "eval_steps_per_second": 0.29,
1451
+ "step": 515
1452
+ },
1453
+ {
1454
+ "epoch": 1.34,
1455
+ "learning_rate": 3.316195372750643e-05,
1456
+ "loss": 1.6354,
1457
+ "step": 520
1458
+ },
1459
+ {
1460
+ "epoch": 1.34,
1461
+ "eval_loss": 1.6610838174819946,
1462
+ "eval_runtime": 37.7265,
1463
+ "eval_samples_per_second": 2.227,
1464
+ "eval_steps_per_second": 0.292,
1465
+ "step": 520
1466
+ },
1467
+ {
1468
+ "epoch": 1.35,
1469
+ "learning_rate": 3.251928020565553e-05,
1470
+ "loss": 1.6201,
1471
+ "step": 525
1472
+ },
1473
+ {
1474
+ "epoch": 1.35,
1475
+ "eval_loss": 1.6610314846038818,
1476
+ "eval_runtime": 37.8464,
1477
+ "eval_samples_per_second": 2.219,
1478
+ "eval_steps_per_second": 0.291,
1479
+ "step": 525
1480
+ },
1481
+ {
1482
+ "epoch": 1.36,
1483
+ "learning_rate": 3.1876606683804625e-05,
1484
+ "loss": 1.6461,
1485
+ "step": 530
1486
+ },
1487
+ {
1488
+ "epoch": 1.36,
1489
+ "eval_loss": 1.6605802774429321,
1490
+ "eval_runtime": 37.9503,
1491
+ "eval_samples_per_second": 2.213,
1492
+ "eval_steps_per_second": 0.29,
1493
+ "step": 530
1494
+ },
1495
+ {
1496
+ "epoch": 1.38,
1497
+ "learning_rate": 3.1233933161953726e-05,
1498
+ "loss": 1.6818,
1499
+ "step": 535
1500
+ },
1501
+ {
1502
+ "epoch": 1.38,
1503
+ "eval_loss": 1.6607571840286255,
1504
+ "eval_runtime": 37.9086,
1505
+ "eval_samples_per_second": 2.216,
1506
+ "eval_steps_per_second": 0.29,
1507
+ "step": 535
1508
+ },
1509
+ {
1510
+ "epoch": 1.39,
1511
+ "learning_rate": 3.059125964010283e-05,
1512
+ "loss": 1.6117,
1513
+ "step": 540
1514
+ },
1515
+ {
1516
+ "epoch": 1.39,
1517
+ "eval_loss": 1.6605336666107178,
1518
+ "eval_runtime": 37.947,
1519
+ "eval_samples_per_second": 2.214,
1520
+ "eval_steps_per_second": 0.29,
1521
+ "step": 540
1522
+ },
1523
+ {
1524
+ "epoch": 1.4,
1525
+ "learning_rate": 2.994858611825193e-05,
1526
+ "loss": 1.6252,
1527
+ "step": 545
1528
+ },
1529
+ {
1530
+ "epoch": 1.4,
1531
+ "eval_loss": 1.6603213548660278,
1532
+ "eval_runtime": 37.9826,
1533
+ "eval_samples_per_second": 2.212,
1534
+ "eval_steps_per_second": 0.29,
1535
+ "step": 545
1536
+ },
1537
+ {
1538
+ "epoch": 1.41,
1539
+ "learning_rate": 2.930591259640103e-05,
1540
+ "loss": 1.6486,
1541
+ "step": 550
1542
+ },
1543
+ {
1544
+ "epoch": 1.41,
1545
+ "eval_loss": 1.6600297689437866,
1546
+ "eval_runtime": 37.9635,
1547
+ "eval_samples_per_second": 2.213,
1548
+ "eval_steps_per_second": 0.29,
1549
+ "step": 550
1550
+ },
1551
+ {
1552
+ "epoch": 1.43,
1553
+ "learning_rate": 2.866323907455013e-05,
1554
+ "loss": 1.6861,
1555
+ "step": 555
1556
+ },
1557
+ {
1558
+ "epoch": 1.43,
1559
+ "eval_loss": 1.6602742671966553,
1560
+ "eval_runtime": 37.8722,
1561
+ "eval_samples_per_second": 2.218,
1562
+ "eval_steps_per_second": 0.29,
1563
+ "step": 555
1564
+ },
1565
+ {
1566
+ "epoch": 1.44,
1567
+ "learning_rate": 2.802056555269923e-05,
1568
+ "loss": 1.6624,
1569
+ "step": 560
1570
+ },
1571
+ {
1572
+ "epoch": 1.44,
1573
+ "eval_loss": 1.6599992513656616,
1574
+ "eval_runtime": 37.9306,
1575
+ "eval_samples_per_second": 2.215,
1576
+ "eval_steps_per_second": 0.29,
1577
+ "step": 560
1578
+ },
1579
+ {
1580
+ "epoch": 1.45,
1581
+ "learning_rate": 2.737789203084833e-05,
1582
+ "loss": 1.6323,
1583
+ "step": 565
1584
+ },
1585
+ {
1586
+ "epoch": 1.45,
1587
+ "eval_loss": 1.6600306034088135,
1588
+ "eval_runtime": 37.9548,
1589
+ "eval_samples_per_second": 2.213,
1590
+ "eval_steps_per_second": 0.29,
1591
+ "step": 565
1592
+ },
1593
+ {
1594
+ "epoch": 1.47,
1595
+ "learning_rate": 2.673521850899743e-05,
1596
+ "loss": 1.6707,
1597
+ "step": 570
1598
+ },
1599
+ {
1600
+ "epoch": 1.47,
1601
+ "eval_loss": 1.6601014137268066,
1602
+ "eval_runtime": 38.0349,
1603
+ "eval_samples_per_second": 2.208,
1604
+ "eval_steps_per_second": 0.289,
1605
+ "step": 570
1606
+ },
1607
+ {
1608
+ "epoch": 1.48,
1609
+ "learning_rate": 2.6092544987146534e-05,
1610
+ "loss": 1.6478,
1611
+ "step": 575
1612
+ },
1613
+ {
1614
+ "epoch": 1.48,
1615
+ "eval_loss": 1.6597967147827148,
1616
+ "eval_runtime": 37.9413,
1617
+ "eval_samples_per_second": 2.214,
1618
+ "eval_steps_per_second": 0.29,
1619
+ "step": 575
1620
+ },
1621
+ {
1622
+ "epoch": 1.49,
1623
+ "learning_rate": 2.5449871465295634e-05,
1624
+ "loss": 1.6715,
1625
+ "step": 580
1626
+ },
1627
+ {
1628
+ "epoch": 1.49,
1629
+ "eval_loss": 1.6598337888717651,
1630
+ "eval_runtime": 37.8386,
1631
+ "eval_samples_per_second": 2.22,
1632
+ "eval_steps_per_second": 0.291,
1633
+ "step": 580
1634
+ },
1635
+ {
1636
+ "epoch": 1.5,
1637
+ "learning_rate": 2.480719794344473e-05,
1638
+ "loss": 1.6626,
1639
+ "step": 585
1640
+ },
1641
+ {
1642
+ "epoch": 1.5,
1643
+ "eval_loss": 1.6600011587142944,
1644
+ "eval_runtime": 37.7901,
1645
+ "eval_samples_per_second": 2.223,
1646
+ "eval_steps_per_second": 0.291,
1647
+ "step": 585
1648
+ },
1649
+ {
1650
+ "epoch": 1.52,
1651
+ "learning_rate": 2.4164524421593832e-05,
1652
+ "loss": 1.651,
1653
+ "step": 590
1654
+ },
1655
+ {
1656
+ "epoch": 1.52,
1657
+ "eval_loss": 1.65969717502594,
1658
+ "eval_runtime": 37.7643,
1659
+ "eval_samples_per_second": 2.224,
1660
+ "eval_steps_per_second": 0.291,
1661
+ "step": 590
1662
+ },
1663
+ {
1664
+ "epoch": 1.53,
1665
+ "learning_rate": 2.3521850899742933e-05,
1666
+ "loss": 1.6346,
1667
+ "step": 595
1668
+ },
1669
+ {
1670
+ "epoch": 1.53,
1671
+ "eval_loss": 1.6598676443099976,
1672
+ "eval_runtime": 37.9026,
1673
+ "eval_samples_per_second": 2.216,
1674
+ "eval_steps_per_second": 0.29,
1675
+ "step": 595
1676
+ },
1677
+ {
1678
+ "epoch": 1.54,
1679
+ "learning_rate": 2.2879177377892033e-05,
1680
+ "loss": 1.6533,
1681
+ "step": 600
1682
+ },
1683
+ {
1684
+ "epoch": 1.54,
1685
+ "eval_loss": 1.659874677658081,
1686
+ "eval_runtime": 37.9871,
1687
+ "eval_samples_per_second": 2.211,
1688
+ "eval_steps_per_second": 0.29,
1689
+ "step": 600
1690
+ },
1691
+ {
1692
+ "epoch": 1.56,
1693
+ "learning_rate": 2.2236503856041134e-05,
1694
+ "loss": 1.6913,
1695
+ "step": 605
1696
+ },
1697
+ {
1698
+ "epoch": 1.56,
1699
+ "eval_loss": 1.659849762916565,
1700
+ "eval_runtime": 37.8712,
1701
+ "eval_samples_per_second": 2.218,
1702
+ "eval_steps_per_second": 0.29,
1703
+ "step": 605
1704
+ },
1705
+ {
1706
+ "epoch": 1.57,
1707
+ "learning_rate": 2.159383033419023e-05,
1708
+ "loss": 1.6709,
1709
+ "step": 610
1710
+ },
1711
+ {
1712
+ "epoch": 1.57,
1713
+ "eval_loss": 1.6598856449127197,
1714
+ "eval_runtime": 37.7043,
1715
+ "eval_samples_per_second": 2.228,
1716
+ "eval_steps_per_second": 0.292,
1717
+ "step": 610
1718
+ },
1719
+ {
1720
+ "epoch": 1.58,
1721
+ "learning_rate": 2.095115681233933e-05,
1722
+ "loss": 1.598,
1723
+ "step": 615
1724
+ },
1725
+ {
1726
+ "epoch": 1.58,
1727
+ "eval_loss": 1.6597983837127686,
1728
+ "eval_runtime": 37.722,
1729
+ "eval_samples_per_second": 2.227,
1730
+ "eval_steps_per_second": 0.292,
1731
+ "step": 615
1732
+ },
1733
+ {
1734
+ "epoch": 1.59,
1735
+ "learning_rate": 2.0308483290488432e-05,
1736
+ "loss": 1.6434,
1737
+ "step": 620
1738
+ },
1739
+ {
1740
+ "epoch": 1.59,
1741
+ "eval_loss": 1.6595433950424194,
1742
+ "eval_runtime": 37.6125,
1743
+ "eval_samples_per_second": 2.233,
1744
+ "eval_steps_per_second": 0.292,
1745
+ "step": 620
1746
+ },
1747
+ {
1748
+ "epoch": 1.61,
1749
+ "learning_rate": 1.9665809768637533e-05,
1750
+ "loss": 1.6086,
1751
+ "step": 625
1752
+ },
1753
+ {
1754
+ "epoch": 1.61,
1755
+ "eval_loss": 1.6594945192337036,
1756
+ "eval_runtime": 38.0046,
1757
+ "eval_samples_per_second": 2.21,
1758
+ "eval_steps_per_second": 0.289,
1759
+ "step": 625
1760
+ },
1761
+ {
1762
+ "epoch": 1.62,
1763
+ "learning_rate": 1.9023136246786633e-05,
1764
+ "loss": 1.5962,
1765
+ "step": 630
1766
+ },
1767
+ {
1768
+ "epoch": 1.62,
1769
+ "eval_loss": 1.6595605611801147,
1770
+ "eval_runtime": 37.8518,
1771
+ "eval_samples_per_second": 2.219,
1772
+ "eval_steps_per_second": 0.291,
1773
+ "step": 630
1774
+ },
1775
+ {
1776
+ "epoch": 1.63,
1777
+ "learning_rate": 1.8380462724935734e-05,
1778
+ "loss": 1.6695,
1779
+ "step": 635
1780
+ },
1781
+ {
1782
+ "epoch": 1.63,
1783
+ "eval_loss": 1.6592403650283813,
1784
+ "eval_runtime": 37.7285,
1785
+ "eval_samples_per_second": 2.226,
1786
+ "eval_steps_per_second": 0.292,
1787
+ "step": 635
1788
+ },
1789
+ {
1790
+ "epoch": 1.65,
1791
+ "learning_rate": 1.7737789203084834e-05,
1792
+ "loss": 1.6526,
1793
+ "step": 640
1794
+ },
1795
+ {
1796
+ "epoch": 1.65,
1797
+ "eval_loss": 1.6591880321502686,
1798
+ "eval_runtime": 37.758,
1799
+ "eval_samples_per_second": 2.225,
1800
+ "eval_steps_per_second": 0.291,
1801
+ "step": 640
1802
+ },
1803
+ {
1804
+ "epoch": 1.66,
1805
+ "learning_rate": 1.7095115681233935e-05,
1806
+ "loss": 1.672,
1807
+ "step": 645
1808
+ },
1809
+ {
1810
+ "epoch": 1.66,
1811
+ "eval_loss": 1.6588138341903687,
1812
+ "eval_runtime": 37.92,
1813
+ "eval_samples_per_second": 2.215,
1814
+ "eval_steps_per_second": 0.29,
1815
+ "step": 645
1816
+ },
1817
+ {
1818
+ "epoch": 1.67,
1819
+ "learning_rate": 1.6452442159383032e-05,
1820
+ "loss": 1.6347,
1821
+ "step": 650
1822
+ },
1823
+ {
1824
+ "epoch": 1.67,
1825
+ "eval_loss": 1.6589809656143188,
1826
+ "eval_runtime": 37.8021,
1827
+ "eval_samples_per_second": 2.222,
1828
+ "eval_steps_per_second": 0.291,
1829
+ "step": 650
1830
+ },
1831
+ {
1832
+ "epoch": 1.68,
1833
+ "learning_rate": 1.5809768637532136e-05,
1834
+ "loss": 1.6227,
1835
+ "step": 655
1836
+ },
1837
+ {
1838
+ "epoch": 1.68,
1839
+ "eval_loss": 1.6586792469024658,
1840
+ "eval_runtime": 37.9128,
1841
+ "eval_samples_per_second": 2.216,
1842
+ "eval_steps_per_second": 0.29,
1843
+ "step": 655
1844
+ },
1845
+ {
1846
+ "epoch": 1.7,
1847
+ "learning_rate": 1.5167095115681235e-05,
1848
+ "loss": 1.6308,
1849
+ "step": 660
1850
+ },
1851
+ {
1852
+ "epoch": 1.7,
1853
+ "eval_loss": 1.6585420370101929,
1854
+ "eval_runtime": 37.9881,
1855
+ "eval_samples_per_second": 2.211,
1856
+ "eval_steps_per_second": 0.29,
1857
+ "step": 660
1858
+ },
1859
+ {
1860
+ "epoch": 1.71,
1861
+ "learning_rate": 1.4524421593830334e-05,
1862
+ "loss": 1.6495,
1863
+ "step": 665
1864
+ },
1865
+ {
1866
+ "epoch": 1.71,
1867
+ "eval_loss": 1.6587530374526978,
1868
+ "eval_runtime": 37.9148,
1869
+ "eval_samples_per_second": 2.215,
1870
+ "eval_steps_per_second": 0.29,
1871
+ "step": 665
1872
+ },
1873
+ {
1874
+ "epoch": 1.72,
1875
+ "learning_rate": 1.3881748071979436e-05,
1876
+ "loss": 1.6712,
1877
+ "step": 670
1878
+ },
1879
+ {
1880
+ "epoch": 1.72,
1881
+ "eval_loss": 1.658648133277893,
1882
+ "eval_runtime": 37.9313,
1883
+ "eval_samples_per_second": 2.215,
1884
+ "eval_steps_per_second": 0.29,
1885
+ "step": 670
1886
+ },
1887
+ {
1888
+ "epoch": 1.74,
1889
+ "learning_rate": 1.3239074550128535e-05,
1890
+ "loss": 1.636,
1891
+ "step": 675
1892
+ },
1893
+ {
1894
+ "epoch": 1.74,
1895
+ "eval_loss": 1.6584407091140747,
1896
+ "eval_runtime": 37.7963,
1897
+ "eval_samples_per_second": 2.222,
1898
+ "eval_steps_per_second": 0.291,
1899
+ "step": 675
1900
+ },
1901
+ {
1902
+ "epoch": 1.75,
1903
+ "learning_rate": 1.2596401028277636e-05,
1904
+ "loss": 1.6453,
1905
+ "step": 680
1906
+ },
1907
+ {
1908
+ "epoch": 1.75,
1909
+ "eval_loss": 1.658594012260437,
1910
+ "eval_runtime": 37.7226,
1911
+ "eval_samples_per_second": 2.227,
1912
+ "eval_steps_per_second": 0.292,
1913
+ "step": 680
1914
+ },
1915
+ {
1916
+ "epoch": 1.76,
1917
+ "learning_rate": 1.1953727506426736e-05,
1918
+ "loss": 1.6509,
1919
+ "step": 685
1920
+ },
1921
+ {
1922
+ "epoch": 1.76,
1923
+ "eval_loss": 1.658467411994934,
1924
+ "eval_runtime": 37.8179,
1925
+ "eval_samples_per_second": 2.221,
1926
+ "eval_steps_per_second": 0.291,
1927
+ "step": 685
1928
+ },
1929
+ {
1930
+ "epoch": 1.77,
1931
+ "learning_rate": 1.1311053984575835e-05,
1932
+ "loss": 1.6489,
1933
+ "step": 690
1934
+ },
1935
+ {
1936
+ "epoch": 1.77,
1937
+ "eval_loss": 1.6585352420806885,
1938
+ "eval_runtime": 37.6768,
1939
+ "eval_samples_per_second": 2.229,
1940
+ "eval_steps_per_second": 0.292,
1941
+ "step": 690
1942
+ },
1943
+ {
1944
+ "epoch": 1.79,
1945
+ "learning_rate": 1.0668380462724936e-05,
1946
+ "loss": 1.6424,
1947
+ "step": 695
1948
+ },
1949
+ {
1950
+ "epoch": 1.79,
1951
+ "eval_loss": 1.6582835912704468,
1952
+ "eval_runtime": 38.0083,
1953
+ "eval_samples_per_second": 2.21,
1954
+ "eval_steps_per_second": 0.289,
1955
+ "step": 695
1956
+ },
1957
+ {
1958
+ "epoch": 1.8,
1959
+ "learning_rate": 1.0025706940874038e-05,
1960
+ "loss": 1.6131,
1961
+ "step": 700
1962
+ },
1963
+ {
1964
+ "epoch": 1.8,
1965
+ "eval_loss": 1.658105731010437,
1966
+ "eval_runtime": 37.8711,
1967
+ "eval_samples_per_second": 2.218,
1968
+ "eval_steps_per_second": 0.29,
1969
+ "step": 700
1970
+ },
1971
+ {
1972
+ "epoch": 1.81,
1973
+ "learning_rate": 9.383033419023137e-06,
1974
+ "loss": 1.6225,
1975
+ "step": 705
1976
+ },
1977
+ {
1978
+ "epoch": 1.81,
1979
+ "eval_loss": 1.658282995223999,
1980
+ "eval_runtime": 37.8125,
1981
+ "eval_samples_per_second": 2.221,
1982
+ "eval_steps_per_second": 0.291,
1983
+ "step": 705
1984
+ },
1985
+ {
1986
+ "epoch": 1.83,
1987
+ "learning_rate": 8.740359897172237e-06,
1988
+ "loss": 1.6928,
1989
+ "step": 710
1990
+ },
1991
+ {
1992
+ "epoch": 1.83,
1993
+ "eval_loss": 1.6581056118011475,
1994
+ "eval_runtime": 37.863,
1995
+ "eval_samples_per_second": 2.219,
1996
+ "eval_steps_per_second": 0.291,
1997
+ "step": 710
1998
+ },
1999
+ {
2000
+ "epoch": 1.84,
2001
+ "learning_rate": 8.097686375321336e-06,
2002
+ "loss": 1.6227,
2003
+ "step": 715
2004
+ },
2005
+ {
2006
+ "epoch": 1.84,
2007
+ "eval_loss": 1.6581188440322876,
2008
+ "eval_runtime": 37.8646,
2009
+ "eval_samples_per_second": 2.218,
2010
+ "eval_steps_per_second": 0.291,
2011
+ "step": 715
2012
+ },
2013
+ {
2014
+ "epoch": 1.85,
2015
+ "learning_rate": 7.4550128534704376e-06,
2016
+ "loss": 1.6593,
2017
+ "step": 720
2018
+ },
2019
+ {
2020
+ "epoch": 1.85,
2021
+ "eval_loss": 1.6579999923706055,
2022
+ "eval_runtime": 37.876,
2023
+ "eval_samples_per_second": 2.218,
2024
+ "eval_steps_per_second": 0.29,
2025
+ "step": 720
2026
+ },
2027
+ {
2028
+ "epoch": 1.86,
2029
+ "learning_rate": 6.812339331619537e-06,
2030
+ "loss": 1.6642,
2031
+ "step": 725
2032
+ },
2033
+ {
2034
+ "epoch": 1.86,
2035
+ "eval_loss": 1.65813410282135,
2036
+ "eval_runtime": 37.7093,
2037
+ "eval_samples_per_second": 2.228,
2038
+ "eval_steps_per_second": 0.292,
2039
+ "step": 725
2040
+ },
2041
+ {
2042
+ "epoch": 1.88,
2043
+ "learning_rate": 6.169665809768638e-06,
2044
+ "loss": 1.6414,
2045
+ "step": 730
2046
+ },
2047
+ {
2048
+ "epoch": 1.88,
2049
+ "eval_loss": 1.6581989526748657,
2050
+ "eval_runtime": 37.7818,
2051
+ "eval_samples_per_second": 2.223,
2052
+ "eval_steps_per_second": 0.291,
2053
+ "step": 730
2054
+ }
2055
+ ],
2056
+ "logging_steps": 5,
2057
+ "max_steps": 778,
2058
+ "num_train_epochs": 2,
2059
+ "save_steps": 10,
2060
+ "total_flos": 2.3714708506804224e+17,
2061
+ "trial_name": null,
2062
+ "trial_params": null
2063
+ }
checkpoint-730/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6211dd8a7344cda681d69cc75786fde8b7049338b76992777b83de12eab260a
3
+ size 4027
checkpoint-740/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: True
10
+ - load_in_4bit: False
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: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.5.0
checkpoint-740/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
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": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-740/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:453309f9dcae7a3d8bafdeedff7735429c075daa092569b7259b1448c1fd0efa
3
+ size 16822989
checkpoint-740/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f806c64f485e0402875f53c800f0ffd84f7a635c3cef973b14cadacbb75b79b3
3
+ size 33665221
checkpoint-740/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9485a713657ce1fde0e837fb7de1440de72c8cd471f9d346ca55fa95f1727688
3
+ size 14575
checkpoint-740/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83cd1616253c229626f1f0350b1e90f3becc16da753c191e13f8f37a45661f5e
3
+ size 627
checkpoint-740/trainer_state.json ADDED
@@ -0,0 +1,2091 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.6579999923706055,
3
+ "best_model_checkpoint": "/scratch/kwamea/llama-output/checkpoint-720",
4
+ "epoch": 1.9023136246786634,
5
+ "eval_steps": 5,
6
+ "global_step": 740,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 9.93573264781491e-05,
14
+ "loss": 1.9824,
15
+ "step": 5
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "eval_loss": 1.954480528831482,
20
+ "eval_runtime": 38.0417,
21
+ "eval_samples_per_second": 2.208,
22
+ "eval_steps_per_second": 0.289,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.03,
27
+ "learning_rate": 9.87146529562982e-05,
28
+ "loss": 1.9245,
29
+ "step": 10
30
+ },
31
+ {
32
+ "epoch": 0.03,
33
+ "eval_loss": 1.8942010402679443,
34
+ "eval_runtime": 37.8934,
35
+ "eval_samples_per_second": 2.217,
36
+ "eval_steps_per_second": 0.29,
37
+ "step": 10
38
+ },
39
+ {
40
+ "epoch": 0.04,
41
+ "learning_rate": 9.80719794344473e-05,
42
+ "loss": 1.8616,
43
+ "step": 15
44
+ },
45
+ {
46
+ "epoch": 0.04,
47
+ "eval_loss": 1.8424055576324463,
48
+ "eval_runtime": 37.8727,
49
+ "eval_samples_per_second": 2.218,
50
+ "eval_steps_per_second": 0.29,
51
+ "step": 15
52
+ },
53
+ {
54
+ "epoch": 0.05,
55
+ "learning_rate": 9.742930591259641e-05,
56
+ "loss": 1.8267,
57
+ "step": 20
58
+ },
59
+ {
60
+ "epoch": 0.05,
61
+ "eval_loss": 1.8279685974121094,
62
+ "eval_runtime": 38.166,
63
+ "eval_samples_per_second": 2.201,
64
+ "eval_steps_per_second": 0.288,
65
+ "step": 20
66
+ },
67
+ {
68
+ "epoch": 0.06,
69
+ "learning_rate": 9.67866323907455e-05,
70
+ "loss": 1.835,
71
+ "step": 25
72
+ },
73
+ {
74
+ "epoch": 0.06,
75
+ "eval_loss": 1.8134318590164185,
76
+ "eval_runtime": 38.0411,
77
+ "eval_samples_per_second": 2.208,
78
+ "eval_steps_per_second": 0.289,
79
+ "step": 25
80
+ },
81
+ {
82
+ "epoch": 0.08,
83
+ "learning_rate": 9.61439588688946e-05,
84
+ "loss": 1.8236,
85
+ "step": 30
86
+ },
87
+ {
88
+ "epoch": 0.08,
89
+ "eval_loss": 1.8010995388031006,
90
+ "eval_runtime": 37.8753,
91
+ "eval_samples_per_second": 2.218,
92
+ "eval_steps_per_second": 0.29,
93
+ "step": 30
94
+ },
95
+ {
96
+ "epoch": 0.09,
97
+ "learning_rate": 9.550128534704372e-05,
98
+ "loss": 1.7186,
99
+ "step": 35
100
+ },
101
+ {
102
+ "epoch": 0.09,
103
+ "eval_loss": 1.7901933193206787,
104
+ "eval_runtime": 38.0374,
105
+ "eval_samples_per_second": 2.208,
106
+ "eval_steps_per_second": 0.289,
107
+ "step": 35
108
+ },
109
+ {
110
+ "epoch": 0.1,
111
+ "learning_rate": 9.485861182519281e-05,
112
+ "loss": 1.7959,
113
+ "step": 40
114
+ },
115
+ {
116
+ "epoch": 0.1,
117
+ "eval_loss": 1.7807564735412598,
118
+ "eval_runtime": 37.8784,
119
+ "eval_samples_per_second": 2.218,
120
+ "eval_steps_per_second": 0.29,
121
+ "step": 40
122
+ },
123
+ {
124
+ "epoch": 0.12,
125
+ "learning_rate": 9.421593830334192e-05,
126
+ "loss": 1.715,
127
+ "step": 45
128
+ },
129
+ {
130
+ "epoch": 0.12,
131
+ "eval_loss": 1.7709146738052368,
132
+ "eval_runtime": 37.8705,
133
+ "eval_samples_per_second": 2.218,
134
+ "eval_steps_per_second": 0.29,
135
+ "step": 45
136
+ },
137
+ {
138
+ "epoch": 0.13,
139
+ "learning_rate": 9.357326478149101e-05,
140
+ "loss": 1.7581,
141
+ "step": 50
142
+ },
143
+ {
144
+ "epoch": 0.13,
145
+ "eval_loss": 1.761339783668518,
146
+ "eval_runtime": 37.661,
147
+ "eval_samples_per_second": 2.23,
148
+ "eval_steps_per_second": 0.292,
149
+ "step": 50
150
+ },
151
+ {
152
+ "epoch": 0.14,
153
+ "learning_rate": 9.29305912596401e-05,
154
+ "loss": 1.7305,
155
+ "step": 55
156
+ },
157
+ {
158
+ "epoch": 0.14,
159
+ "eval_loss": 1.7389689683914185,
160
+ "eval_runtime": 38.0055,
161
+ "eval_samples_per_second": 2.21,
162
+ "eval_steps_per_second": 0.289,
163
+ "step": 55
164
+ },
165
+ {
166
+ "epoch": 0.15,
167
+ "learning_rate": 9.228791773778921e-05,
168
+ "loss": 1.7086,
169
+ "step": 60
170
+ },
171
+ {
172
+ "epoch": 0.15,
173
+ "eval_loss": 1.713218092918396,
174
+ "eval_runtime": 37.737,
175
+ "eval_samples_per_second": 2.226,
176
+ "eval_steps_per_second": 0.291,
177
+ "step": 60
178
+ },
179
+ {
180
+ "epoch": 0.17,
181
+ "learning_rate": 9.16452442159383e-05,
182
+ "loss": 1.7057,
183
+ "step": 65
184
+ },
185
+ {
186
+ "epoch": 0.17,
187
+ "eval_loss": 1.7108122110366821,
188
+ "eval_runtime": 38.0037,
189
+ "eval_samples_per_second": 2.21,
190
+ "eval_steps_per_second": 0.289,
191
+ "step": 65
192
+ },
193
+ {
194
+ "epoch": 0.18,
195
+ "learning_rate": 9.100257069408741e-05,
196
+ "loss": 1.7243,
197
+ "step": 70
198
+ },
199
+ {
200
+ "epoch": 0.18,
201
+ "eval_loss": 1.7041622400283813,
202
+ "eval_runtime": 38.1402,
203
+ "eval_samples_per_second": 2.202,
204
+ "eval_steps_per_second": 0.288,
205
+ "step": 70
206
+ },
207
+ {
208
+ "epoch": 0.19,
209
+ "learning_rate": 9.03598971722365e-05,
210
+ "loss": 1.7038,
211
+ "step": 75
212
+ },
213
+ {
214
+ "epoch": 0.19,
215
+ "eval_loss": 1.7010408639907837,
216
+ "eval_runtime": 37.7193,
217
+ "eval_samples_per_second": 2.227,
218
+ "eval_steps_per_second": 0.292,
219
+ "step": 75
220
+ },
221
+ {
222
+ "epoch": 0.21,
223
+ "learning_rate": 8.97172236503856e-05,
224
+ "loss": 1.6939,
225
+ "step": 80
226
+ },
227
+ {
228
+ "epoch": 0.21,
229
+ "eval_loss": 1.6987501382827759,
230
+ "eval_runtime": 38.0624,
231
+ "eval_samples_per_second": 2.207,
232
+ "eval_steps_per_second": 0.289,
233
+ "step": 80
234
+ },
235
+ {
236
+ "epoch": 0.22,
237
+ "learning_rate": 8.907455012853471e-05,
238
+ "loss": 1.645,
239
+ "step": 85
240
+ },
241
+ {
242
+ "epoch": 0.22,
243
+ "eval_loss": 1.6968218088150024,
244
+ "eval_runtime": 37.9205,
245
+ "eval_samples_per_second": 2.215,
246
+ "eval_steps_per_second": 0.29,
247
+ "step": 85
248
+ },
249
+ {
250
+ "epoch": 0.23,
251
+ "learning_rate": 8.84318766066838e-05,
252
+ "loss": 1.7016,
253
+ "step": 90
254
+ },
255
+ {
256
+ "epoch": 0.23,
257
+ "eval_loss": 1.6952500343322754,
258
+ "eval_runtime": 37.9617,
259
+ "eval_samples_per_second": 2.213,
260
+ "eval_steps_per_second": 0.29,
261
+ "step": 90
262
+ },
263
+ {
264
+ "epoch": 0.24,
265
+ "learning_rate": 8.778920308483291e-05,
266
+ "loss": 1.6869,
267
+ "step": 95
268
+ },
269
+ {
270
+ "epoch": 0.24,
271
+ "eval_loss": 1.6932854652404785,
272
+ "eval_runtime": 37.8781,
273
+ "eval_samples_per_second": 2.218,
274
+ "eval_steps_per_second": 0.29,
275
+ "step": 95
276
+ },
277
+ {
278
+ "epoch": 0.26,
279
+ "learning_rate": 8.7146529562982e-05,
280
+ "loss": 1.6728,
281
+ "step": 100
282
+ },
283
+ {
284
+ "epoch": 0.26,
285
+ "eval_loss": 1.6920738220214844,
286
+ "eval_runtime": 38.1996,
287
+ "eval_samples_per_second": 2.199,
288
+ "eval_steps_per_second": 0.288,
289
+ "step": 100
290
+ },
291
+ {
292
+ "epoch": 0.27,
293
+ "learning_rate": 8.650385604113111e-05,
294
+ "loss": 1.6674,
295
+ "step": 105
296
+ },
297
+ {
298
+ "epoch": 0.27,
299
+ "eval_loss": 1.6904469728469849,
300
+ "eval_runtime": 38.1252,
301
+ "eval_samples_per_second": 2.203,
302
+ "eval_steps_per_second": 0.289,
303
+ "step": 105
304
+ },
305
+ {
306
+ "epoch": 0.28,
307
+ "learning_rate": 8.586118251928022e-05,
308
+ "loss": 1.7013,
309
+ "step": 110
310
+ },
311
+ {
312
+ "epoch": 0.28,
313
+ "eval_loss": 1.6895124912261963,
314
+ "eval_runtime": 38.0968,
315
+ "eval_samples_per_second": 2.205,
316
+ "eval_steps_per_second": 0.289,
317
+ "step": 110
318
+ },
319
+ {
320
+ "epoch": 0.3,
321
+ "learning_rate": 8.521850899742931e-05,
322
+ "loss": 1.7148,
323
+ "step": 115
324
+ },
325
+ {
326
+ "epoch": 0.3,
327
+ "eval_loss": 1.6886374950408936,
328
+ "eval_runtime": 37.8799,
329
+ "eval_samples_per_second": 2.218,
330
+ "eval_steps_per_second": 0.29,
331
+ "step": 115
332
+ },
333
+ {
334
+ "epoch": 0.31,
335
+ "learning_rate": 8.457583547557842e-05,
336
+ "loss": 1.7166,
337
+ "step": 120
338
+ },
339
+ {
340
+ "epoch": 0.31,
341
+ "eval_loss": 1.6868164539337158,
342
+ "eval_runtime": 37.8861,
343
+ "eval_samples_per_second": 2.217,
344
+ "eval_steps_per_second": 0.29,
345
+ "step": 120
346
+ },
347
+ {
348
+ "epoch": 0.32,
349
+ "learning_rate": 8.393316195372751e-05,
350
+ "loss": 1.7012,
351
+ "step": 125
352
+ },
353
+ {
354
+ "epoch": 0.32,
355
+ "eval_loss": 1.6858367919921875,
356
+ "eval_runtime": 38.1136,
357
+ "eval_samples_per_second": 2.204,
358
+ "eval_steps_per_second": 0.289,
359
+ "step": 125
360
+ },
361
+ {
362
+ "epoch": 0.33,
363
+ "learning_rate": 8.32904884318766e-05,
364
+ "loss": 1.6827,
365
+ "step": 130
366
+ },
367
+ {
368
+ "epoch": 0.33,
369
+ "eval_loss": 1.6849361658096313,
370
+ "eval_runtime": 37.9883,
371
+ "eval_samples_per_second": 2.211,
372
+ "eval_steps_per_second": 0.29,
373
+ "step": 130
374
+ },
375
+ {
376
+ "epoch": 0.35,
377
+ "learning_rate": 8.264781491002571e-05,
378
+ "loss": 1.7157,
379
+ "step": 135
380
+ },
381
+ {
382
+ "epoch": 0.35,
383
+ "eval_loss": 1.684756875038147,
384
+ "eval_runtime": 37.8726,
385
+ "eval_samples_per_second": 2.218,
386
+ "eval_steps_per_second": 0.29,
387
+ "step": 135
388
+ },
389
+ {
390
+ "epoch": 0.36,
391
+ "learning_rate": 8.200514138817481e-05,
392
+ "loss": 1.6668,
393
+ "step": 140
394
+ },
395
+ {
396
+ "epoch": 0.36,
397
+ "eval_loss": 1.6836236715316772,
398
+ "eval_runtime": 37.9586,
399
+ "eval_samples_per_second": 2.213,
400
+ "eval_steps_per_second": 0.29,
401
+ "step": 140
402
+ },
403
+ {
404
+ "epoch": 0.37,
405
+ "learning_rate": 8.136246786632391e-05,
406
+ "loss": 1.6588,
407
+ "step": 145
408
+ },
409
+ {
410
+ "epoch": 0.37,
411
+ "eval_loss": 1.6828166246414185,
412
+ "eval_runtime": 38.0596,
413
+ "eval_samples_per_second": 2.207,
414
+ "eval_steps_per_second": 0.289,
415
+ "step": 145
416
+ },
417
+ {
418
+ "epoch": 0.39,
419
+ "learning_rate": 8.071979434447301e-05,
420
+ "loss": 1.7005,
421
+ "step": 150
422
+ },
423
+ {
424
+ "epoch": 0.39,
425
+ "eval_loss": 1.681463360786438,
426
+ "eval_runtime": 38.0878,
427
+ "eval_samples_per_second": 2.205,
428
+ "eval_steps_per_second": 0.289,
429
+ "step": 150
430
+ },
431
+ {
432
+ "epoch": 0.4,
433
+ "learning_rate": 8.007712082262212e-05,
434
+ "loss": 1.6893,
435
+ "step": 155
436
+ },
437
+ {
438
+ "epoch": 0.4,
439
+ "eval_loss": 1.6811552047729492,
440
+ "eval_runtime": 37.9265,
441
+ "eval_samples_per_second": 2.215,
442
+ "eval_steps_per_second": 0.29,
443
+ "step": 155
444
+ },
445
+ {
446
+ "epoch": 0.41,
447
+ "learning_rate": 7.943444730077121e-05,
448
+ "loss": 1.6682,
449
+ "step": 160
450
+ },
451
+ {
452
+ "epoch": 0.41,
453
+ "eval_loss": 1.6801096200942993,
454
+ "eval_runtime": 38.0375,
455
+ "eval_samples_per_second": 2.208,
456
+ "eval_steps_per_second": 0.289,
457
+ "step": 160
458
+ },
459
+ {
460
+ "epoch": 0.42,
461
+ "learning_rate": 7.87917737789203e-05,
462
+ "loss": 1.6481,
463
+ "step": 165
464
+ },
465
+ {
466
+ "epoch": 0.42,
467
+ "eval_loss": 1.6800144910812378,
468
+ "eval_runtime": 37.8916,
469
+ "eval_samples_per_second": 2.217,
470
+ "eval_steps_per_second": 0.29,
471
+ "step": 165
472
+ },
473
+ {
474
+ "epoch": 0.44,
475
+ "learning_rate": 7.814910025706941e-05,
476
+ "loss": 1.7042,
477
+ "step": 170
478
+ },
479
+ {
480
+ "epoch": 0.44,
481
+ "eval_loss": 1.6784976720809937,
482
+ "eval_runtime": 37.8514,
483
+ "eval_samples_per_second": 2.219,
484
+ "eval_steps_per_second": 0.291,
485
+ "step": 170
486
+ },
487
+ {
488
+ "epoch": 0.45,
489
+ "learning_rate": 7.750642673521852e-05,
490
+ "loss": 1.6555,
491
+ "step": 175
492
+ },
493
+ {
494
+ "epoch": 0.45,
495
+ "eval_loss": 1.6780468225479126,
496
+ "eval_runtime": 37.8893,
497
+ "eval_samples_per_second": 2.217,
498
+ "eval_steps_per_second": 0.29,
499
+ "step": 175
500
+ },
501
+ {
502
+ "epoch": 0.46,
503
+ "learning_rate": 7.686375321336761e-05,
504
+ "loss": 1.6717,
505
+ "step": 180
506
+ },
507
+ {
508
+ "epoch": 0.46,
509
+ "eval_loss": 1.6775085926055908,
510
+ "eval_runtime": 37.9289,
511
+ "eval_samples_per_second": 2.215,
512
+ "eval_steps_per_second": 0.29,
513
+ "step": 180
514
+ },
515
+ {
516
+ "epoch": 0.48,
517
+ "learning_rate": 7.622107969151672e-05,
518
+ "loss": 1.6716,
519
+ "step": 185
520
+ },
521
+ {
522
+ "epoch": 0.48,
523
+ "eval_loss": 1.6770914793014526,
524
+ "eval_runtime": 38.0111,
525
+ "eval_samples_per_second": 2.21,
526
+ "eval_steps_per_second": 0.289,
527
+ "step": 185
528
+ },
529
+ {
530
+ "epoch": 0.49,
531
+ "learning_rate": 7.557840616966581e-05,
532
+ "loss": 1.6618,
533
+ "step": 190
534
+ },
535
+ {
536
+ "epoch": 0.49,
537
+ "eval_loss": 1.676563024520874,
538
+ "eval_runtime": 38.0703,
539
+ "eval_samples_per_second": 2.206,
540
+ "eval_steps_per_second": 0.289,
541
+ "step": 190
542
+ },
543
+ {
544
+ "epoch": 0.5,
545
+ "learning_rate": 7.493573264781492e-05,
546
+ "loss": 1.6411,
547
+ "step": 195
548
+ },
549
+ {
550
+ "epoch": 0.5,
551
+ "eval_loss": 1.6757832765579224,
552
+ "eval_runtime": 37.827,
553
+ "eval_samples_per_second": 2.221,
554
+ "eval_steps_per_second": 0.291,
555
+ "step": 195
556
+ },
557
+ {
558
+ "epoch": 0.51,
559
+ "learning_rate": 7.429305912596401e-05,
560
+ "loss": 1.6637,
561
+ "step": 200
562
+ },
563
+ {
564
+ "epoch": 0.51,
565
+ "eval_loss": 1.6751985549926758,
566
+ "eval_runtime": 37.836,
567
+ "eval_samples_per_second": 2.22,
568
+ "eval_steps_per_second": 0.291,
569
+ "step": 200
570
+ },
571
+ {
572
+ "epoch": 0.53,
573
+ "learning_rate": 7.365038560411311e-05,
574
+ "loss": 1.6142,
575
+ "step": 205
576
+ },
577
+ {
578
+ "epoch": 0.53,
579
+ "eval_loss": 1.675147294998169,
580
+ "eval_runtime": 37.9175,
581
+ "eval_samples_per_second": 2.215,
582
+ "eval_steps_per_second": 0.29,
583
+ "step": 205
584
+ },
585
+ {
586
+ "epoch": 0.54,
587
+ "learning_rate": 7.300771208226222e-05,
588
+ "loss": 1.6741,
589
+ "step": 210
590
+ },
591
+ {
592
+ "epoch": 0.54,
593
+ "eval_loss": 1.674437165260315,
594
+ "eval_runtime": 38.087,
595
+ "eval_samples_per_second": 2.205,
596
+ "eval_steps_per_second": 0.289,
597
+ "step": 210
598
+ },
599
+ {
600
+ "epoch": 0.55,
601
+ "learning_rate": 7.236503856041131e-05,
602
+ "loss": 1.6408,
603
+ "step": 215
604
+ },
605
+ {
606
+ "epoch": 0.55,
607
+ "eval_loss": 1.6737719774246216,
608
+ "eval_runtime": 38.0075,
609
+ "eval_samples_per_second": 2.21,
610
+ "eval_steps_per_second": 0.289,
611
+ "step": 215
612
+ },
613
+ {
614
+ "epoch": 0.57,
615
+ "learning_rate": 7.172236503856042e-05,
616
+ "loss": 1.6733,
617
+ "step": 220
618
+ },
619
+ {
620
+ "epoch": 0.57,
621
+ "eval_loss": 1.6732759475708008,
622
+ "eval_runtime": 38.1125,
623
+ "eval_samples_per_second": 2.204,
624
+ "eval_steps_per_second": 0.289,
625
+ "step": 220
626
+ },
627
+ {
628
+ "epoch": 0.58,
629
+ "learning_rate": 7.107969151670951e-05,
630
+ "loss": 1.679,
631
+ "step": 225
632
+ },
633
+ {
634
+ "epoch": 0.58,
635
+ "eval_loss": 1.6726195812225342,
636
+ "eval_runtime": 37.9769,
637
+ "eval_samples_per_second": 2.212,
638
+ "eval_steps_per_second": 0.29,
639
+ "step": 225
640
+ },
641
+ {
642
+ "epoch": 0.59,
643
+ "learning_rate": 7.043701799485862e-05,
644
+ "loss": 1.7202,
645
+ "step": 230
646
+ },
647
+ {
648
+ "epoch": 0.59,
649
+ "eval_loss": 1.671908974647522,
650
+ "eval_runtime": 37.8215,
651
+ "eval_samples_per_second": 2.221,
652
+ "eval_steps_per_second": 0.291,
653
+ "step": 230
654
+ },
655
+ {
656
+ "epoch": 0.6,
657
+ "learning_rate": 6.979434447300771e-05,
658
+ "loss": 1.6805,
659
+ "step": 235
660
+ },
661
+ {
662
+ "epoch": 0.6,
663
+ "eval_loss": 1.6715577840805054,
664
+ "eval_runtime": 37.9343,
665
+ "eval_samples_per_second": 2.214,
666
+ "eval_steps_per_second": 0.29,
667
+ "step": 235
668
+ },
669
+ {
670
+ "epoch": 0.62,
671
+ "learning_rate": 6.91516709511568e-05,
672
+ "loss": 1.6331,
673
+ "step": 240
674
+ },
675
+ {
676
+ "epoch": 0.62,
677
+ "eval_loss": 1.6715376377105713,
678
+ "eval_runtime": 37.9904,
679
+ "eval_samples_per_second": 2.211,
680
+ "eval_steps_per_second": 0.29,
681
+ "step": 240
682
+ },
683
+ {
684
+ "epoch": 0.63,
685
+ "learning_rate": 6.850899742930593e-05,
686
+ "loss": 1.6761,
687
+ "step": 245
688
+ },
689
+ {
690
+ "epoch": 0.63,
691
+ "eval_loss": 1.671446681022644,
692
+ "eval_runtime": 38.0357,
693
+ "eval_samples_per_second": 2.208,
694
+ "eval_steps_per_second": 0.289,
695
+ "step": 245
696
+ },
697
+ {
698
+ "epoch": 0.64,
699
+ "learning_rate": 6.786632390745502e-05,
700
+ "loss": 1.6994,
701
+ "step": 250
702
+ },
703
+ {
704
+ "epoch": 0.64,
705
+ "eval_loss": 1.6716604232788086,
706
+ "eval_runtime": 37.9609,
707
+ "eval_samples_per_second": 2.213,
708
+ "eval_steps_per_second": 0.29,
709
+ "step": 250
710
+ },
711
+ {
712
+ "epoch": 0.66,
713
+ "learning_rate": 6.722365038560411e-05,
714
+ "loss": 1.6305,
715
+ "step": 255
716
+ },
717
+ {
718
+ "epoch": 0.66,
719
+ "eval_loss": 1.6711723804473877,
720
+ "eval_runtime": 37.7893,
721
+ "eval_samples_per_second": 2.223,
722
+ "eval_steps_per_second": 0.291,
723
+ "step": 255
724
+ },
725
+ {
726
+ "epoch": 0.67,
727
+ "learning_rate": 6.658097686375322e-05,
728
+ "loss": 1.6612,
729
+ "step": 260
730
+ },
731
+ {
732
+ "epoch": 0.67,
733
+ "eval_loss": 1.670398473739624,
734
+ "eval_runtime": 37.7667,
735
+ "eval_samples_per_second": 2.224,
736
+ "eval_steps_per_second": 0.291,
737
+ "step": 260
738
+ },
739
+ {
740
+ "epoch": 0.68,
741
+ "learning_rate": 6.593830334190231e-05,
742
+ "loss": 1.6576,
743
+ "step": 265
744
+ },
745
+ {
746
+ "epoch": 0.68,
747
+ "eval_loss": 1.6706459522247314,
748
+ "eval_runtime": 37.8285,
749
+ "eval_samples_per_second": 2.221,
750
+ "eval_steps_per_second": 0.291,
751
+ "step": 265
752
+ },
753
+ {
754
+ "epoch": 0.69,
755
+ "learning_rate": 6.529562982005142e-05,
756
+ "loss": 1.6837,
757
+ "step": 270
758
+ },
759
+ {
760
+ "epoch": 0.69,
761
+ "eval_loss": 1.6699285507202148,
762
+ "eval_runtime": 37.7619,
763
+ "eval_samples_per_second": 2.224,
764
+ "eval_steps_per_second": 0.291,
765
+ "step": 270
766
+ },
767
+ {
768
+ "epoch": 0.71,
769
+ "learning_rate": 6.465295629820052e-05,
770
+ "loss": 1.6493,
771
+ "step": 275
772
+ },
773
+ {
774
+ "epoch": 0.71,
775
+ "eval_loss": 1.6693135499954224,
776
+ "eval_runtime": 37.7441,
777
+ "eval_samples_per_second": 2.226,
778
+ "eval_steps_per_second": 0.291,
779
+ "step": 275
780
+ },
781
+ {
782
+ "epoch": 0.72,
783
+ "learning_rate": 6.401028277634962e-05,
784
+ "loss": 1.6427,
785
+ "step": 280
786
+ },
787
+ {
788
+ "epoch": 0.72,
789
+ "eval_loss": 1.669467568397522,
790
+ "eval_runtime": 37.8707,
791
+ "eval_samples_per_second": 2.218,
792
+ "eval_steps_per_second": 0.29,
793
+ "step": 280
794
+ },
795
+ {
796
+ "epoch": 0.73,
797
+ "learning_rate": 6.336760925449872e-05,
798
+ "loss": 1.6236,
799
+ "step": 285
800
+ },
801
+ {
802
+ "epoch": 0.73,
803
+ "eval_loss": 1.6690431833267212,
804
+ "eval_runtime": 37.8946,
805
+ "eval_samples_per_second": 2.217,
806
+ "eval_steps_per_second": 0.29,
807
+ "step": 285
808
+ },
809
+ {
810
+ "epoch": 0.75,
811
+ "learning_rate": 6.272493573264781e-05,
812
+ "loss": 1.6538,
813
+ "step": 290
814
+ },
815
+ {
816
+ "epoch": 0.75,
817
+ "eval_loss": 1.6688117980957031,
818
+ "eval_runtime": 37.9378,
819
+ "eval_samples_per_second": 2.214,
820
+ "eval_steps_per_second": 0.29,
821
+ "step": 290
822
+ },
823
+ {
824
+ "epoch": 0.76,
825
+ "learning_rate": 6.208226221079692e-05,
826
+ "loss": 1.6558,
827
+ "step": 295
828
+ },
829
+ {
830
+ "epoch": 0.76,
831
+ "eval_loss": 1.668560266494751,
832
+ "eval_runtime": 38.0448,
833
+ "eval_samples_per_second": 2.208,
834
+ "eval_steps_per_second": 0.289,
835
+ "step": 295
836
+ },
837
+ {
838
+ "epoch": 0.77,
839
+ "learning_rate": 6.143958868894601e-05,
840
+ "loss": 1.6401,
841
+ "step": 300
842
+ },
843
+ {
844
+ "epoch": 0.77,
845
+ "eval_loss": 1.6680612564086914,
846
+ "eval_runtime": 37.8571,
847
+ "eval_samples_per_second": 2.219,
848
+ "eval_steps_per_second": 0.291,
849
+ "step": 300
850
+ },
851
+ {
852
+ "epoch": 0.78,
853
+ "learning_rate": 6.079691516709511e-05,
854
+ "loss": 1.6468,
855
+ "step": 305
856
+ },
857
+ {
858
+ "epoch": 0.78,
859
+ "eval_loss": 1.6675914525985718,
860
+ "eval_runtime": 37.8491,
861
+ "eval_samples_per_second": 2.219,
862
+ "eval_steps_per_second": 0.291,
863
+ "step": 305
864
+ },
865
+ {
866
+ "epoch": 0.8,
867
+ "learning_rate": 6.015424164524421e-05,
868
+ "loss": 1.6579,
869
+ "step": 310
870
+ },
871
+ {
872
+ "epoch": 0.8,
873
+ "eval_loss": 1.6671632528305054,
874
+ "eval_runtime": 37.7658,
875
+ "eval_samples_per_second": 2.224,
876
+ "eval_steps_per_second": 0.291,
877
+ "step": 310
878
+ },
879
+ {
880
+ "epoch": 0.81,
881
+ "learning_rate": 5.951156812339333e-05,
882
+ "loss": 1.6339,
883
+ "step": 315
884
+ },
885
+ {
886
+ "epoch": 0.81,
887
+ "eval_loss": 1.666803240776062,
888
+ "eval_runtime": 37.7896,
889
+ "eval_samples_per_second": 2.223,
890
+ "eval_steps_per_second": 0.291,
891
+ "step": 315
892
+ },
893
+ {
894
+ "epoch": 0.82,
895
+ "learning_rate": 5.886889460154242e-05,
896
+ "loss": 1.6636,
897
+ "step": 320
898
+ },
899
+ {
900
+ "epoch": 0.82,
901
+ "eval_loss": 1.6664105653762817,
902
+ "eval_runtime": 38.0586,
903
+ "eval_samples_per_second": 2.207,
904
+ "eval_steps_per_second": 0.289,
905
+ "step": 320
906
+ },
907
+ {
908
+ "epoch": 0.84,
909
+ "learning_rate": 5.822622107969152e-05,
910
+ "loss": 1.641,
911
+ "step": 325
912
+ },
913
+ {
914
+ "epoch": 0.84,
915
+ "eval_loss": 1.6665875911712646,
916
+ "eval_runtime": 37.9046,
917
+ "eval_samples_per_second": 2.216,
918
+ "eval_steps_per_second": 0.29,
919
+ "step": 325
920
+ },
921
+ {
922
+ "epoch": 0.85,
923
+ "learning_rate": 5.758354755784062e-05,
924
+ "loss": 1.6616,
925
+ "step": 330
926
+ },
927
+ {
928
+ "epoch": 0.85,
929
+ "eval_loss": 1.6664165258407593,
930
+ "eval_runtime": 37.8575,
931
+ "eval_samples_per_second": 2.219,
932
+ "eval_steps_per_second": 0.291,
933
+ "step": 330
934
+ },
935
+ {
936
+ "epoch": 0.86,
937
+ "learning_rate": 5.694087403598972e-05,
938
+ "loss": 1.6597,
939
+ "step": 335
940
+ },
941
+ {
942
+ "epoch": 0.86,
943
+ "eval_loss": 1.666045904159546,
944
+ "eval_runtime": 37.9711,
945
+ "eval_samples_per_second": 2.212,
946
+ "eval_steps_per_second": 0.29,
947
+ "step": 335
948
+ },
949
+ {
950
+ "epoch": 0.87,
951
+ "learning_rate": 5.6298200514138824e-05,
952
+ "loss": 1.67,
953
+ "step": 340
954
+ },
955
+ {
956
+ "epoch": 0.87,
957
+ "eval_loss": 1.666337490081787,
958
+ "eval_runtime": 37.8821,
959
+ "eval_samples_per_second": 2.217,
960
+ "eval_steps_per_second": 0.29,
961
+ "step": 340
962
+ },
963
+ {
964
+ "epoch": 0.89,
965
+ "learning_rate": 5.5655526992287924e-05,
966
+ "loss": 1.6344,
967
+ "step": 345
968
+ },
969
+ {
970
+ "epoch": 0.89,
971
+ "eval_loss": 1.6659029722213745,
972
+ "eval_runtime": 37.9188,
973
+ "eval_samples_per_second": 2.215,
974
+ "eval_steps_per_second": 0.29,
975
+ "step": 345
976
+ },
977
+ {
978
+ "epoch": 0.9,
979
+ "learning_rate": 5.501285347043702e-05,
980
+ "loss": 1.6623,
981
+ "step": 350
982
+ },
983
+ {
984
+ "epoch": 0.9,
985
+ "eval_loss": 1.6656525135040283,
986
+ "eval_runtime": 37.9951,
987
+ "eval_samples_per_second": 2.211,
988
+ "eval_steps_per_second": 0.29,
989
+ "step": 350
990
+ },
991
+ {
992
+ "epoch": 0.91,
993
+ "learning_rate": 5.437017994858612e-05,
994
+ "loss": 1.6623,
995
+ "step": 355
996
+ },
997
+ {
998
+ "epoch": 0.91,
999
+ "eval_loss": 1.6654421091079712,
1000
+ "eval_runtime": 37.8803,
1001
+ "eval_samples_per_second": 2.218,
1002
+ "eval_steps_per_second": 0.29,
1003
+ "step": 355
1004
+ },
1005
+ {
1006
+ "epoch": 0.93,
1007
+ "learning_rate": 5.372750642673522e-05,
1008
+ "loss": 1.6741,
1009
+ "step": 360
1010
+ },
1011
+ {
1012
+ "epoch": 0.93,
1013
+ "eval_loss": 1.665541648864746,
1014
+ "eval_runtime": 37.7151,
1015
+ "eval_samples_per_second": 2.227,
1016
+ "eval_steps_per_second": 0.292,
1017
+ "step": 360
1018
+ },
1019
+ {
1020
+ "epoch": 0.94,
1021
+ "learning_rate": 5.308483290488432e-05,
1022
+ "loss": 1.6973,
1023
+ "step": 365
1024
+ },
1025
+ {
1026
+ "epoch": 0.94,
1027
+ "eval_loss": 1.6651471853256226,
1028
+ "eval_runtime": 37.9144,
1029
+ "eval_samples_per_second": 2.216,
1030
+ "eval_steps_per_second": 0.29,
1031
+ "step": 365
1032
+ },
1033
+ {
1034
+ "epoch": 0.95,
1035
+ "learning_rate": 5.244215938303342e-05,
1036
+ "loss": 1.6829,
1037
+ "step": 370
1038
+ },
1039
+ {
1040
+ "epoch": 0.95,
1041
+ "eval_loss": 1.6651064157485962,
1042
+ "eval_runtime": 37.7443,
1043
+ "eval_samples_per_second": 2.226,
1044
+ "eval_steps_per_second": 0.291,
1045
+ "step": 370
1046
+ },
1047
+ {
1048
+ "epoch": 0.96,
1049
+ "learning_rate": 5.1799485861182514e-05,
1050
+ "loss": 1.6923,
1051
+ "step": 375
1052
+ },
1053
+ {
1054
+ "epoch": 0.96,
1055
+ "eval_loss": 1.6646034717559814,
1056
+ "eval_runtime": 37.8783,
1057
+ "eval_samples_per_second": 2.218,
1058
+ "eval_steps_per_second": 0.29,
1059
+ "step": 375
1060
+ },
1061
+ {
1062
+ "epoch": 0.98,
1063
+ "learning_rate": 5.1156812339331615e-05,
1064
+ "loss": 1.6725,
1065
+ "step": 380
1066
+ },
1067
+ {
1068
+ "epoch": 0.98,
1069
+ "eval_loss": 1.6639235019683838,
1070
+ "eval_runtime": 37.7559,
1071
+ "eval_samples_per_second": 2.225,
1072
+ "eval_steps_per_second": 0.291,
1073
+ "step": 380
1074
+ },
1075
+ {
1076
+ "epoch": 0.99,
1077
+ "learning_rate": 5.051413881748073e-05,
1078
+ "loss": 1.6216,
1079
+ "step": 385
1080
+ },
1081
+ {
1082
+ "epoch": 0.99,
1083
+ "eval_loss": 1.6643506288528442,
1084
+ "eval_runtime": 37.9895,
1085
+ "eval_samples_per_second": 2.211,
1086
+ "eval_steps_per_second": 0.29,
1087
+ "step": 385
1088
+ },
1089
+ {
1090
+ "epoch": 1.0,
1091
+ "learning_rate": 4.987146529562982e-05,
1092
+ "loss": 1.6518,
1093
+ "step": 390
1094
+ },
1095
+ {
1096
+ "epoch": 1.0,
1097
+ "eval_loss": 1.6643046140670776,
1098
+ "eval_runtime": 37.9289,
1099
+ "eval_samples_per_second": 2.215,
1100
+ "eval_steps_per_second": 0.29,
1101
+ "step": 390
1102
+ },
1103
+ {
1104
+ "epoch": 1.02,
1105
+ "learning_rate": 4.922879177377892e-05,
1106
+ "loss": 1.658,
1107
+ "step": 395
1108
+ },
1109
+ {
1110
+ "epoch": 1.02,
1111
+ "eval_loss": 1.6641957759857178,
1112
+ "eval_runtime": 37.8638,
1113
+ "eval_samples_per_second": 2.218,
1114
+ "eval_steps_per_second": 0.291,
1115
+ "step": 395
1116
+ },
1117
+ {
1118
+ "epoch": 1.03,
1119
+ "learning_rate": 4.8586118251928024e-05,
1120
+ "loss": 1.6767,
1121
+ "step": 400
1122
+ },
1123
+ {
1124
+ "epoch": 1.03,
1125
+ "eval_loss": 1.663757085800171,
1126
+ "eval_runtime": 37.7728,
1127
+ "eval_samples_per_second": 2.224,
1128
+ "eval_steps_per_second": 0.291,
1129
+ "step": 400
1130
+ },
1131
+ {
1132
+ "epoch": 1.04,
1133
+ "learning_rate": 4.7943444730077124e-05,
1134
+ "loss": 1.6264,
1135
+ "step": 405
1136
+ },
1137
+ {
1138
+ "epoch": 1.04,
1139
+ "eval_loss": 1.6635041236877441,
1140
+ "eval_runtime": 37.7702,
1141
+ "eval_samples_per_second": 2.224,
1142
+ "eval_steps_per_second": 0.291,
1143
+ "step": 405
1144
+ },
1145
+ {
1146
+ "epoch": 1.05,
1147
+ "learning_rate": 4.7300771208226225e-05,
1148
+ "loss": 1.6527,
1149
+ "step": 410
1150
+ },
1151
+ {
1152
+ "epoch": 1.05,
1153
+ "eval_loss": 1.6632280349731445,
1154
+ "eval_runtime": 37.8453,
1155
+ "eval_samples_per_second": 2.22,
1156
+ "eval_steps_per_second": 0.291,
1157
+ "step": 410
1158
+ },
1159
+ {
1160
+ "epoch": 1.07,
1161
+ "learning_rate": 4.6658097686375325e-05,
1162
+ "loss": 1.6157,
1163
+ "step": 415
1164
+ },
1165
+ {
1166
+ "epoch": 1.07,
1167
+ "eval_loss": 1.6636598110198975,
1168
+ "eval_runtime": 37.812,
1169
+ "eval_samples_per_second": 2.222,
1170
+ "eval_steps_per_second": 0.291,
1171
+ "step": 415
1172
+ },
1173
+ {
1174
+ "epoch": 1.08,
1175
+ "learning_rate": 4.6015424164524426e-05,
1176
+ "loss": 1.5966,
1177
+ "step": 420
1178
+ },
1179
+ {
1180
+ "epoch": 1.08,
1181
+ "eval_loss": 1.663393497467041,
1182
+ "eval_runtime": 37.7743,
1183
+ "eval_samples_per_second": 2.224,
1184
+ "eval_steps_per_second": 0.291,
1185
+ "step": 420
1186
+ },
1187
+ {
1188
+ "epoch": 1.09,
1189
+ "learning_rate": 4.537275064267352e-05,
1190
+ "loss": 1.6705,
1191
+ "step": 425
1192
+ },
1193
+ {
1194
+ "epoch": 1.09,
1195
+ "eval_loss": 1.6631269454956055,
1196
+ "eval_runtime": 38.0953,
1197
+ "eval_samples_per_second": 2.205,
1198
+ "eval_steps_per_second": 0.289,
1199
+ "step": 425
1200
+ },
1201
+ {
1202
+ "epoch": 1.11,
1203
+ "learning_rate": 4.473007712082262e-05,
1204
+ "loss": 1.6691,
1205
+ "step": 430
1206
+ },
1207
+ {
1208
+ "epoch": 1.11,
1209
+ "eval_loss": 1.6633110046386719,
1210
+ "eval_runtime": 37.8964,
1211
+ "eval_samples_per_second": 2.217,
1212
+ "eval_steps_per_second": 0.29,
1213
+ "step": 430
1214
+ },
1215
+ {
1216
+ "epoch": 1.12,
1217
+ "learning_rate": 4.408740359897173e-05,
1218
+ "loss": 1.6332,
1219
+ "step": 435
1220
+ },
1221
+ {
1222
+ "epoch": 1.12,
1223
+ "eval_loss": 1.6628649234771729,
1224
+ "eval_runtime": 38.0621,
1225
+ "eval_samples_per_second": 2.207,
1226
+ "eval_steps_per_second": 0.289,
1227
+ "step": 435
1228
+ },
1229
+ {
1230
+ "epoch": 1.13,
1231
+ "learning_rate": 4.344473007712083e-05,
1232
+ "loss": 1.5916,
1233
+ "step": 440
1234
+ },
1235
+ {
1236
+ "epoch": 1.13,
1237
+ "eval_loss": 1.662834882736206,
1238
+ "eval_runtime": 37.9722,
1239
+ "eval_samples_per_second": 2.212,
1240
+ "eval_steps_per_second": 0.29,
1241
+ "step": 440
1242
+ },
1243
+ {
1244
+ "epoch": 1.14,
1245
+ "learning_rate": 4.280205655526993e-05,
1246
+ "loss": 1.6543,
1247
+ "step": 445
1248
+ },
1249
+ {
1250
+ "epoch": 1.14,
1251
+ "eval_loss": 1.6630265712738037,
1252
+ "eval_runtime": 37.9699,
1253
+ "eval_samples_per_second": 2.212,
1254
+ "eval_steps_per_second": 0.29,
1255
+ "step": 445
1256
+ },
1257
+ {
1258
+ "epoch": 1.16,
1259
+ "learning_rate": 4.215938303341902e-05,
1260
+ "loss": 1.6353,
1261
+ "step": 450
1262
+ },
1263
+ {
1264
+ "epoch": 1.16,
1265
+ "eval_loss": 1.6625033617019653,
1266
+ "eval_runtime": 37.7505,
1267
+ "eval_samples_per_second": 2.225,
1268
+ "eval_steps_per_second": 0.291,
1269
+ "step": 450
1270
+ },
1271
+ {
1272
+ "epoch": 1.17,
1273
+ "learning_rate": 4.151670951156812e-05,
1274
+ "loss": 1.6441,
1275
+ "step": 455
1276
+ },
1277
+ {
1278
+ "epoch": 1.17,
1279
+ "eval_loss": 1.662785530090332,
1280
+ "eval_runtime": 37.75,
1281
+ "eval_samples_per_second": 2.225,
1282
+ "eval_steps_per_second": 0.291,
1283
+ "step": 455
1284
+ },
1285
+ {
1286
+ "epoch": 1.18,
1287
+ "learning_rate": 4.0874035989717224e-05,
1288
+ "loss": 1.6631,
1289
+ "step": 460
1290
+ },
1291
+ {
1292
+ "epoch": 1.18,
1293
+ "eval_loss": 1.6627779006958008,
1294
+ "eval_runtime": 37.864,
1295
+ "eval_samples_per_second": 2.218,
1296
+ "eval_steps_per_second": 0.291,
1297
+ "step": 460
1298
+ },
1299
+ {
1300
+ "epoch": 1.2,
1301
+ "learning_rate": 4.0231362467866324e-05,
1302
+ "loss": 1.6327,
1303
+ "step": 465
1304
+ },
1305
+ {
1306
+ "epoch": 1.2,
1307
+ "eval_loss": 1.662458896636963,
1308
+ "eval_runtime": 37.7496,
1309
+ "eval_samples_per_second": 2.225,
1310
+ "eval_steps_per_second": 0.291,
1311
+ "step": 465
1312
+ },
1313
+ {
1314
+ "epoch": 1.21,
1315
+ "learning_rate": 3.958868894601543e-05,
1316
+ "loss": 1.6238,
1317
+ "step": 470
1318
+ },
1319
+ {
1320
+ "epoch": 1.21,
1321
+ "eval_loss": 1.6621876955032349,
1322
+ "eval_runtime": 37.8943,
1323
+ "eval_samples_per_second": 2.217,
1324
+ "eval_steps_per_second": 0.29,
1325
+ "step": 470
1326
+ },
1327
+ {
1328
+ "epoch": 1.22,
1329
+ "learning_rate": 3.8946015424164526e-05,
1330
+ "loss": 1.6231,
1331
+ "step": 475
1332
+ },
1333
+ {
1334
+ "epoch": 1.22,
1335
+ "eval_loss": 1.662013292312622,
1336
+ "eval_runtime": 37.9364,
1337
+ "eval_samples_per_second": 2.214,
1338
+ "eval_steps_per_second": 0.29,
1339
+ "step": 475
1340
+ },
1341
+ {
1342
+ "epoch": 1.23,
1343
+ "learning_rate": 3.8303341902313626e-05,
1344
+ "loss": 1.6381,
1345
+ "step": 480
1346
+ },
1347
+ {
1348
+ "epoch": 1.23,
1349
+ "eval_loss": 1.6615785360336304,
1350
+ "eval_runtime": 37.91,
1351
+ "eval_samples_per_second": 2.216,
1352
+ "eval_steps_per_second": 0.29,
1353
+ "step": 480
1354
+ },
1355
+ {
1356
+ "epoch": 1.25,
1357
+ "learning_rate": 3.766066838046273e-05,
1358
+ "loss": 1.6419,
1359
+ "step": 485
1360
+ },
1361
+ {
1362
+ "epoch": 1.25,
1363
+ "eval_loss": 1.662042498588562,
1364
+ "eval_runtime": 37.7834,
1365
+ "eval_samples_per_second": 2.223,
1366
+ "eval_steps_per_second": 0.291,
1367
+ "step": 485
1368
+ },
1369
+ {
1370
+ "epoch": 1.26,
1371
+ "learning_rate": 3.701799485861183e-05,
1372
+ "loss": 1.6319,
1373
+ "step": 490
1374
+ },
1375
+ {
1376
+ "epoch": 1.26,
1377
+ "eval_loss": 1.6619195938110352,
1378
+ "eval_runtime": 37.7385,
1379
+ "eval_samples_per_second": 2.226,
1380
+ "eval_steps_per_second": 0.291,
1381
+ "step": 490
1382
+ },
1383
+ {
1384
+ "epoch": 1.27,
1385
+ "learning_rate": 3.637532133676093e-05,
1386
+ "loss": 1.609,
1387
+ "step": 495
1388
+ },
1389
+ {
1390
+ "epoch": 1.27,
1391
+ "eval_loss": 1.6615225076675415,
1392
+ "eval_runtime": 37.8056,
1393
+ "eval_samples_per_second": 2.222,
1394
+ "eval_steps_per_second": 0.291,
1395
+ "step": 495
1396
+ },
1397
+ {
1398
+ "epoch": 1.29,
1399
+ "learning_rate": 3.573264781491003e-05,
1400
+ "loss": 1.6435,
1401
+ "step": 500
1402
+ },
1403
+ {
1404
+ "epoch": 1.29,
1405
+ "eval_loss": 1.6610548496246338,
1406
+ "eval_runtime": 37.9353,
1407
+ "eval_samples_per_second": 2.214,
1408
+ "eval_steps_per_second": 0.29,
1409
+ "step": 500
1410
+ },
1411
+ {
1412
+ "epoch": 1.3,
1413
+ "learning_rate": 3.508997429305913e-05,
1414
+ "loss": 1.6381,
1415
+ "step": 505
1416
+ },
1417
+ {
1418
+ "epoch": 1.3,
1419
+ "eval_loss": 1.661049246788025,
1420
+ "eval_runtime": 37.9669,
1421
+ "eval_samples_per_second": 2.212,
1422
+ "eval_steps_per_second": 0.29,
1423
+ "step": 505
1424
+ },
1425
+ {
1426
+ "epoch": 1.31,
1427
+ "learning_rate": 3.444730077120823e-05,
1428
+ "loss": 1.6506,
1429
+ "step": 510
1430
+ },
1431
+ {
1432
+ "epoch": 1.31,
1433
+ "eval_loss": 1.6610089540481567,
1434
+ "eval_runtime": 37.9127,
1435
+ "eval_samples_per_second": 2.216,
1436
+ "eval_steps_per_second": 0.29,
1437
+ "step": 510
1438
+ },
1439
+ {
1440
+ "epoch": 1.32,
1441
+ "learning_rate": 3.380462724935733e-05,
1442
+ "loss": 1.6376,
1443
+ "step": 515
1444
+ },
1445
+ {
1446
+ "epoch": 1.32,
1447
+ "eval_loss": 1.6608952283859253,
1448
+ "eval_runtime": 37.9454,
1449
+ "eval_samples_per_second": 2.214,
1450
+ "eval_steps_per_second": 0.29,
1451
+ "step": 515
1452
+ },
1453
+ {
1454
+ "epoch": 1.34,
1455
+ "learning_rate": 3.316195372750643e-05,
1456
+ "loss": 1.6354,
1457
+ "step": 520
1458
+ },
1459
+ {
1460
+ "epoch": 1.34,
1461
+ "eval_loss": 1.6610838174819946,
1462
+ "eval_runtime": 37.7265,
1463
+ "eval_samples_per_second": 2.227,
1464
+ "eval_steps_per_second": 0.292,
1465
+ "step": 520
1466
+ },
1467
+ {
1468
+ "epoch": 1.35,
1469
+ "learning_rate": 3.251928020565553e-05,
1470
+ "loss": 1.6201,
1471
+ "step": 525
1472
+ },
1473
+ {
1474
+ "epoch": 1.35,
1475
+ "eval_loss": 1.6610314846038818,
1476
+ "eval_runtime": 37.8464,
1477
+ "eval_samples_per_second": 2.219,
1478
+ "eval_steps_per_second": 0.291,
1479
+ "step": 525
1480
+ },
1481
+ {
1482
+ "epoch": 1.36,
1483
+ "learning_rate": 3.1876606683804625e-05,
1484
+ "loss": 1.6461,
1485
+ "step": 530
1486
+ },
1487
+ {
1488
+ "epoch": 1.36,
1489
+ "eval_loss": 1.6605802774429321,
1490
+ "eval_runtime": 37.9503,
1491
+ "eval_samples_per_second": 2.213,
1492
+ "eval_steps_per_second": 0.29,
1493
+ "step": 530
1494
+ },
1495
+ {
1496
+ "epoch": 1.38,
1497
+ "learning_rate": 3.1233933161953726e-05,
1498
+ "loss": 1.6818,
1499
+ "step": 535
1500
+ },
1501
+ {
1502
+ "epoch": 1.38,
1503
+ "eval_loss": 1.6607571840286255,
1504
+ "eval_runtime": 37.9086,
1505
+ "eval_samples_per_second": 2.216,
1506
+ "eval_steps_per_second": 0.29,
1507
+ "step": 535
1508
+ },
1509
+ {
1510
+ "epoch": 1.39,
1511
+ "learning_rate": 3.059125964010283e-05,
1512
+ "loss": 1.6117,
1513
+ "step": 540
1514
+ },
1515
+ {
1516
+ "epoch": 1.39,
1517
+ "eval_loss": 1.6605336666107178,
1518
+ "eval_runtime": 37.947,
1519
+ "eval_samples_per_second": 2.214,
1520
+ "eval_steps_per_second": 0.29,
1521
+ "step": 540
1522
+ },
1523
+ {
1524
+ "epoch": 1.4,
1525
+ "learning_rate": 2.994858611825193e-05,
1526
+ "loss": 1.6252,
1527
+ "step": 545
1528
+ },
1529
+ {
1530
+ "epoch": 1.4,
1531
+ "eval_loss": 1.6603213548660278,
1532
+ "eval_runtime": 37.9826,
1533
+ "eval_samples_per_second": 2.212,
1534
+ "eval_steps_per_second": 0.29,
1535
+ "step": 545
1536
+ },
1537
+ {
1538
+ "epoch": 1.41,
1539
+ "learning_rate": 2.930591259640103e-05,
1540
+ "loss": 1.6486,
1541
+ "step": 550
1542
+ },
1543
+ {
1544
+ "epoch": 1.41,
1545
+ "eval_loss": 1.6600297689437866,
1546
+ "eval_runtime": 37.9635,
1547
+ "eval_samples_per_second": 2.213,
1548
+ "eval_steps_per_second": 0.29,
1549
+ "step": 550
1550
+ },
1551
+ {
1552
+ "epoch": 1.43,
1553
+ "learning_rate": 2.866323907455013e-05,
1554
+ "loss": 1.6861,
1555
+ "step": 555
1556
+ },
1557
+ {
1558
+ "epoch": 1.43,
1559
+ "eval_loss": 1.6602742671966553,
1560
+ "eval_runtime": 37.8722,
1561
+ "eval_samples_per_second": 2.218,
1562
+ "eval_steps_per_second": 0.29,
1563
+ "step": 555
1564
+ },
1565
+ {
1566
+ "epoch": 1.44,
1567
+ "learning_rate": 2.802056555269923e-05,
1568
+ "loss": 1.6624,
1569
+ "step": 560
1570
+ },
1571
+ {
1572
+ "epoch": 1.44,
1573
+ "eval_loss": 1.6599992513656616,
1574
+ "eval_runtime": 37.9306,
1575
+ "eval_samples_per_second": 2.215,
1576
+ "eval_steps_per_second": 0.29,
1577
+ "step": 560
1578
+ },
1579
+ {
1580
+ "epoch": 1.45,
1581
+ "learning_rate": 2.737789203084833e-05,
1582
+ "loss": 1.6323,
1583
+ "step": 565
1584
+ },
1585
+ {
1586
+ "epoch": 1.45,
1587
+ "eval_loss": 1.6600306034088135,
1588
+ "eval_runtime": 37.9548,
1589
+ "eval_samples_per_second": 2.213,
1590
+ "eval_steps_per_second": 0.29,
1591
+ "step": 565
1592
+ },
1593
+ {
1594
+ "epoch": 1.47,
1595
+ "learning_rate": 2.673521850899743e-05,
1596
+ "loss": 1.6707,
1597
+ "step": 570
1598
+ },
1599
+ {
1600
+ "epoch": 1.47,
1601
+ "eval_loss": 1.6601014137268066,
1602
+ "eval_runtime": 38.0349,
1603
+ "eval_samples_per_second": 2.208,
1604
+ "eval_steps_per_second": 0.289,
1605
+ "step": 570
1606
+ },
1607
+ {
1608
+ "epoch": 1.48,
1609
+ "learning_rate": 2.6092544987146534e-05,
1610
+ "loss": 1.6478,
1611
+ "step": 575
1612
+ },
1613
+ {
1614
+ "epoch": 1.48,
1615
+ "eval_loss": 1.6597967147827148,
1616
+ "eval_runtime": 37.9413,
1617
+ "eval_samples_per_second": 2.214,
1618
+ "eval_steps_per_second": 0.29,
1619
+ "step": 575
1620
+ },
1621
+ {
1622
+ "epoch": 1.49,
1623
+ "learning_rate": 2.5449871465295634e-05,
1624
+ "loss": 1.6715,
1625
+ "step": 580
1626
+ },
1627
+ {
1628
+ "epoch": 1.49,
1629
+ "eval_loss": 1.6598337888717651,
1630
+ "eval_runtime": 37.8386,
1631
+ "eval_samples_per_second": 2.22,
1632
+ "eval_steps_per_second": 0.291,
1633
+ "step": 580
1634
+ },
1635
+ {
1636
+ "epoch": 1.5,
1637
+ "learning_rate": 2.480719794344473e-05,
1638
+ "loss": 1.6626,
1639
+ "step": 585
1640
+ },
1641
+ {
1642
+ "epoch": 1.5,
1643
+ "eval_loss": 1.6600011587142944,
1644
+ "eval_runtime": 37.7901,
1645
+ "eval_samples_per_second": 2.223,
1646
+ "eval_steps_per_second": 0.291,
1647
+ "step": 585
1648
+ },
1649
+ {
1650
+ "epoch": 1.52,
1651
+ "learning_rate": 2.4164524421593832e-05,
1652
+ "loss": 1.651,
1653
+ "step": 590
1654
+ },
1655
+ {
1656
+ "epoch": 1.52,
1657
+ "eval_loss": 1.65969717502594,
1658
+ "eval_runtime": 37.7643,
1659
+ "eval_samples_per_second": 2.224,
1660
+ "eval_steps_per_second": 0.291,
1661
+ "step": 590
1662
+ },
1663
+ {
1664
+ "epoch": 1.53,
1665
+ "learning_rate": 2.3521850899742933e-05,
1666
+ "loss": 1.6346,
1667
+ "step": 595
1668
+ },
1669
+ {
1670
+ "epoch": 1.53,
1671
+ "eval_loss": 1.6598676443099976,
1672
+ "eval_runtime": 37.9026,
1673
+ "eval_samples_per_second": 2.216,
1674
+ "eval_steps_per_second": 0.29,
1675
+ "step": 595
1676
+ },
1677
+ {
1678
+ "epoch": 1.54,
1679
+ "learning_rate": 2.2879177377892033e-05,
1680
+ "loss": 1.6533,
1681
+ "step": 600
1682
+ },
1683
+ {
1684
+ "epoch": 1.54,
1685
+ "eval_loss": 1.659874677658081,
1686
+ "eval_runtime": 37.9871,
1687
+ "eval_samples_per_second": 2.211,
1688
+ "eval_steps_per_second": 0.29,
1689
+ "step": 600
1690
+ },
1691
+ {
1692
+ "epoch": 1.56,
1693
+ "learning_rate": 2.2236503856041134e-05,
1694
+ "loss": 1.6913,
1695
+ "step": 605
1696
+ },
1697
+ {
1698
+ "epoch": 1.56,
1699
+ "eval_loss": 1.659849762916565,
1700
+ "eval_runtime": 37.8712,
1701
+ "eval_samples_per_second": 2.218,
1702
+ "eval_steps_per_second": 0.29,
1703
+ "step": 605
1704
+ },
1705
+ {
1706
+ "epoch": 1.57,
1707
+ "learning_rate": 2.159383033419023e-05,
1708
+ "loss": 1.6709,
1709
+ "step": 610
1710
+ },
1711
+ {
1712
+ "epoch": 1.57,
1713
+ "eval_loss": 1.6598856449127197,
1714
+ "eval_runtime": 37.7043,
1715
+ "eval_samples_per_second": 2.228,
1716
+ "eval_steps_per_second": 0.292,
1717
+ "step": 610
1718
+ },
1719
+ {
1720
+ "epoch": 1.58,
1721
+ "learning_rate": 2.095115681233933e-05,
1722
+ "loss": 1.598,
1723
+ "step": 615
1724
+ },
1725
+ {
1726
+ "epoch": 1.58,
1727
+ "eval_loss": 1.6597983837127686,
1728
+ "eval_runtime": 37.722,
1729
+ "eval_samples_per_second": 2.227,
1730
+ "eval_steps_per_second": 0.292,
1731
+ "step": 615
1732
+ },
1733
+ {
1734
+ "epoch": 1.59,
1735
+ "learning_rate": 2.0308483290488432e-05,
1736
+ "loss": 1.6434,
1737
+ "step": 620
1738
+ },
1739
+ {
1740
+ "epoch": 1.59,
1741
+ "eval_loss": 1.6595433950424194,
1742
+ "eval_runtime": 37.6125,
1743
+ "eval_samples_per_second": 2.233,
1744
+ "eval_steps_per_second": 0.292,
1745
+ "step": 620
1746
+ },
1747
+ {
1748
+ "epoch": 1.61,
1749
+ "learning_rate": 1.9665809768637533e-05,
1750
+ "loss": 1.6086,
1751
+ "step": 625
1752
+ },
1753
+ {
1754
+ "epoch": 1.61,
1755
+ "eval_loss": 1.6594945192337036,
1756
+ "eval_runtime": 38.0046,
1757
+ "eval_samples_per_second": 2.21,
1758
+ "eval_steps_per_second": 0.289,
1759
+ "step": 625
1760
+ },
1761
+ {
1762
+ "epoch": 1.62,
1763
+ "learning_rate": 1.9023136246786633e-05,
1764
+ "loss": 1.5962,
1765
+ "step": 630
1766
+ },
1767
+ {
1768
+ "epoch": 1.62,
1769
+ "eval_loss": 1.6595605611801147,
1770
+ "eval_runtime": 37.8518,
1771
+ "eval_samples_per_second": 2.219,
1772
+ "eval_steps_per_second": 0.291,
1773
+ "step": 630
1774
+ },
1775
+ {
1776
+ "epoch": 1.63,
1777
+ "learning_rate": 1.8380462724935734e-05,
1778
+ "loss": 1.6695,
1779
+ "step": 635
1780
+ },
1781
+ {
1782
+ "epoch": 1.63,
1783
+ "eval_loss": 1.6592403650283813,
1784
+ "eval_runtime": 37.7285,
1785
+ "eval_samples_per_second": 2.226,
1786
+ "eval_steps_per_second": 0.292,
1787
+ "step": 635
1788
+ },
1789
+ {
1790
+ "epoch": 1.65,
1791
+ "learning_rate": 1.7737789203084834e-05,
1792
+ "loss": 1.6526,
1793
+ "step": 640
1794
+ },
1795
+ {
1796
+ "epoch": 1.65,
1797
+ "eval_loss": 1.6591880321502686,
1798
+ "eval_runtime": 37.758,
1799
+ "eval_samples_per_second": 2.225,
1800
+ "eval_steps_per_second": 0.291,
1801
+ "step": 640
1802
+ },
1803
+ {
1804
+ "epoch": 1.66,
1805
+ "learning_rate": 1.7095115681233935e-05,
1806
+ "loss": 1.672,
1807
+ "step": 645
1808
+ },
1809
+ {
1810
+ "epoch": 1.66,
1811
+ "eval_loss": 1.6588138341903687,
1812
+ "eval_runtime": 37.92,
1813
+ "eval_samples_per_second": 2.215,
1814
+ "eval_steps_per_second": 0.29,
1815
+ "step": 645
1816
+ },
1817
+ {
1818
+ "epoch": 1.67,
1819
+ "learning_rate": 1.6452442159383032e-05,
1820
+ "loss": 1.6347,
1821
+ "step": 650
1822
+ },
1823
+ {
1824
+ "epoch": 1.67,
1825
+ "eval_loss": 1.6589809656143188,
1826
+ "eval_runtime": 37.8021,
1827
+ "eval_samples_per_second": 2.222,
1828
+ "eval_steps_per_second": 0.291,
1829
+ "step": 650
1830
+ },
1831
+ {
1832
+ "epoch": 1.68,
1833
+ "learning_rate": 1.5809768637532136e-05,
1834
+ "loss": 1.6227,
1835
+ "step": 655
1836
+ },
1837
+ {
1838
+ "epoch": 1.68,
1839
+ "eval_loss": 1.6586792469024658,
1840
+ "eval_runtime": 37.9128,
1841
+ "eval_samples_per_second": 2.216,
1842
+ "eval_steps_per_second": 0.29,
1843
+ "step": 655
1844
+ },
1845
+ {
1846
+ "epoch": 1.7,
1847
+ "learning_rate": 1.5167095115681235e-05,
1848
+ "loss": 1.6308,
1849
+ "step": 660
1850
+ },
1851
+ {
1852
+ "epoch": 1.7,
1853
+ "eval_loss": 1.6585420370101929,
1854
+ "eval_runtime": 37.9881,
1855
+ "eval_samples_per_second": 2.211,
1856
+ "eval_steps_per_second": 0.29,
1857
+ "step": 660
1858
+ },
1859
+ {
1860
+ "epoch": 1.71,
1861
+ "learning_rate": 1.4524421593830334e-05,
1862
+ "loss": 1.6495,
1863
+ "step": 665
1864
+ },
1865
+ {
1866
+ "epoch": 1.71,
1867
+ "eval_loss": 1.6587530374526978,
1868
+ "eval_runtime": 37.9148,
1869
+ "eval_samples_per_second": 2.215,
1870
+ "eval_steps_per_second": 0.29,
1871
+ "step": 665
1872
+ },
1873
+ {
1874
+ "epoch": 1.72,
1875
+ "learning_rate": 1.3881748071979436e-05,
1876
+ "loss": 1.6712,
1877
+ "step": 670
1878
+ },
1879
+ {
1880
+ "epoch": 1.72,
1881
+ "eval_loss": 1.658648133277893,
1882
+ "eval_runtime": 37.9313,
1883
+ "eval_samples_per_second": 2.215,
1884
+ "eval_steps_per_second": 0.29,
1885
+ "step": 670
1886
+ },
1887
+ {
1888
+ "epoch": 1.74,
1889
+ "learning_rate": 1.3239074550128535e-05,
1890
+ "loss": 1.636,
1891
+ "step": 675
1892
+ },
1893
+ {
1894
+ "epoch": 1.74,
1895
+ "eval_loss": 1.6584407091140747,
1896
+ "eval_runtime": 37.7963,
1897
+ "eval_samples_per_second": 2.222,
1898
+ "eval_steps_per_second": 0.291,
1899
+ "step": 675
1900
+ },
1901
+ {
1902
+ "epoch": 1.75,
1903
+ "learning_rate": 1.2596401028277636e-05,
1904
+ "loss": 1.6453,
1905
+ "step": 680
1906
+ },
1907
+ {
1908
+ "epoch": 1.75,
1909
+ "eval_loss": 1.658594012260437,
1910
+ "eval_runtime": 37.7226,
1911
+ "eval_samples_per_second": 2.227,
1912
+ "eval_steps_per_second": 0.292,
1913
+ "step": 680
1914
+ },
1915
+ {
1916
+ "epoch": 1.76,
1917
+ "learning_rate": 1.1953727506426736e-05,
1918
+ "loss": 1.6509,
1919
+ "step": 685
1920
+ },
1921
+ {
1922
+ "epoch": 1.76,
1923
+ "eval_loss": 1.658467411994934,
1924
+ "eval_runtime": 37.8179,
1925
+ "eval_samples_per_second": 2.221,
1926
+ "eval_steps_per_second": 0.291,
1927
+ "step": 685
1928
+ },
1929
+ {
1930
+ "epoch": 1.77,
1931
+ "learning_rate": 1.1311053984575835e-05,
1932
+ "loss": 1.6489,
1933
+ "step": 690
1934
+ },
1935
+ {
1936
+ "epoch": 1.77,
1937
+ "eval_loss": 1.6585352420806885,
1938
+ "eval_runtime": 37.6768,
1939
+ "eval_samples_per_second": 2.229,
1940
+ "eval_steps_per_second": 0.292,
1941
+ "step": 690
1942
+ },
1943
+ {
1944
+ "epoch": 1.79,
1945
+ "learning_rate": 1.0668380462724936e-05,
1946
+ "loss": 1.6424,
1947
+ "step": 695
1948
+ },
1949
+ {
1950
+ "epoch": 1.79,
1951
+ "eval_loss": 1.6582835912704468,
1952
+ "eval_runtime": 38.0083,
1953
+ "eval_samples_per_second": 2.21,
1954
+ "eval_steps_per_second": 0.289,
1955
+ "step": 695
1956
+ },
1957
+ {
1958
+ "epoch": 1.8,
1959
+ "learning_rate": 1.0025706940874038e-05,
1960
+ "loss": 1.6131,
1961
+ "step": 700
1962
+ },
1963
+ {
1964
+ "epoch": 1.8,
1965
+ "eval_loss": 1.658105731010437,
1966
+ "eval_runtime": 37.8711,
1967
+ "eval_samples_per_second": 2.218,
1968
+ "eval_steps_per_second": 0.29,
1969
+ "step": 700
1970
+ },
1971
+ {
1972
+ "epoch": 1.81,
1973
+ "learning_rate": 9.383033419023137e-06,
1974
+ "loss": 1.6225,
1975
+ "step": 705
1976
+ },
1977
+ {
1978
+ "epoch": 1.81,
1979
+ "eval_loss": 1.658282995223999,
1980
+ "eval_runtime": 37.8125,
1981
+ "eval_samples_per_second": 2.221,
1982
+ "eval_steps_per_second": 0.291,
1983
+ "step": 705
1984
+ },
1985
+ {
1986
+ "epoch": 1.83,
1987
+ "learning_rate": 8.740359897172237e-06,
1988
+ "loss": 1.6928,
1989
+ "step": 710
1990
+ },
1991
+ {
1992
+ "epoch": 1.83,
1993
+ "eval_loss": 1.6581056118011475,
1994
+ "eval_runtime": 37.863,
1995
+ "eval_samples_per_second": 2.219,
1996
+ "eval_steps_per_second": 0.291,
1997
+ "step": 710
1998
+ },
1999
+ {
2000
+ "epoch": 1.84,
2001
+ "learning_rate": 8.097686375321336e-06,
2002
+ "loss": 1.6227,
2003
+ "step": 715
2004
+ },
2005
+ {
2006
+ "epoch": 1.84,
2007
+ "eval_loss": 1.6581188440322876,
2008
+ "eval_runtime": 37.8646,
2009
+ "eval_samples_per_second": 2.218,
2010
+ "eval_steps_per_second": 0.291,
2011
+ "step": 715
2012
+ },
2013
+ {
2014
+ "epoch": 1.85,
2015
+ "learning_rate": 7.4550128534704376e-06,
2016
+ "loss": 1.6593,
2017
+ "step": 720
2018
+ },
2019
+ {
2020
+ "epoch": 1.85,
2021
+ "eval_loss": 1.6579999923706055,
2022
+ "eval_runtime": 37.876,
2023
+ "eval_samples_per_second": 2.218,
2024
+ "eval_steps_per_second": 0.29,
2025
+ "step": 720
2026
+ },
2027
+ {
2028
+ "epoch": 1.86,
2029
+ "learning_rate": 6.812339331619537e-06,
2030
+ "loss": 1.6642,
2031
+ "step": 725
2032
+ },
2033
+ {
2034
+ "epoch": 1.86,
2035
+ "eval_loss": 1.65813410282135,
2036
+ "eval_runtime": 37.7093,
2037
+ "eval_samples_per_second": 2.228,
2038
+ "eval_steps_per_second": 0.292,
2039
+ "step": 725
2040
+ },
2041
+ {
2042
+ "epoch": 1.88,
2043
+ "learning_rate": 6.169665809768638e-06,
2044
+ "loss": 1.6414,
2045
+ "step": 730
2046
+ },
2047
+ {
2048
+ "epoch": 1.88,
2049
+ "eval_loss": 1.6581989526748657,
2050
+ "eval_runtime": 37.7818,
2051
+ "eval_samples_per_second": 2.223,
2052
+ "eval_steps_per_second": 0.291,
2053
+ "step": 730
2054
+ },
2055
+ {
2056
+ "epoch": 1.89,
2057
+ "learning_rate": 5.526992287917738e-06,
2058
+ "loss": 1.6519,
2059
+ "step": 735
2060
+ },
2061
+ {
2062
+ "epoch": 1.89,
2063
+ "eval_loss": 1.6577394008636475,
2064
+ "eval_runtime": 37.7264,
2065
+ "eval_samples_per_second": 2.227,
2066
+ "eval_steps_per_second": 0.292,
2067
+ "step": 735
2068
+ },
2069
+ {
2070
+ "epoch": 1.9,
2071
+ "learning_rate": 4.884318766066838e-06,
2072
+ "loss": 1.6485,
2073
+ "step": 740
2074
+ },
2075
+ {
2076
+ "epoch": 1.9,
2077
+ "eval_loss": 1.6580238342285156,
2078
+ "eval_runtime": 37.8729,
2079
+ "eval_samples_per_second": 2.218,
2080
+ "eval_steps_per_second": 0.29,
2081
+ "step": 740
2082
+ }
2083
+ ],
2084
+ "logging_steps": 5,
2085
+ "max_steps": 778,
2086
+ "num_train_epochs": 2,
2087
+ "save_steps": 10,
2088
+ "total_flos": 2.4039678818648064e+17,
2089
+ "trial_name": null,
2090
+ "trial_params": null
2091
+ }
checkpoint-740/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6211dd8a7344cda681d69cc75786fde8b7049338b76992777b83de12eab260a
3
+ size 4027
checkpoint-750/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: True
10
+ - load_in_4bit: False
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: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.5.0
checkpoint-750/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
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": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-750/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c0d660f3ab33615be4f5c1d611774912fb7aec65ea170b3d85ad3e852ca7fac
3
+ size 16822989
checkpoint-750/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dda819b3f1710011ce3f08254a4bb9ff0e65c2a49e979920f834b4a3d8965247
3
+ size 33665221
checkpoint-750/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac7c8f0b470b9a83502a647540c7e0ade02b17dc033f116150a8cc2e4dc6aacd
3
+ size 14575
checkpoint-750/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f25f594dd23bb3702632a3e26270da7703d0eb2831a6db8d72406f9c7ffeddf2
3
+ size 627
checkpoint-750/trainer_state.json ADDED
@@ -0,0 +1,2119 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.657867670059204,
3
+ "best_model_checkpoint": "/scratch/kwamea/llama-output/checkpoint-750",
4
+ "epoch": 1.9280205655526992,
5
+ "eval_steps": 5,
6
+ "global_step": 750,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 9.93573264781491e-05,
14
+ "loss": 1.9824,
15
+ "step": 5
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "eval_loss": 1.954480528831482,
20
+ "eval_runtime": 38.0417,
21
+ "eval_samples_per_second": 2.208,
22
+ "eval_steps_per_second": 0.289,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.03,
27
+ "learning_rate": 9.87146529562982e-05,
28
+ "loss": 1.9245,
29
+ "step": 10
30
+ },
31
+ {
32
+ "epoch": 0.03,
33
+ "eval_loss": 1.8942010402679443,
34
+ "eval_runtime": 37.8934,
35
+ "eval_samples_per_second": 2.217,
36
+ "eval_steps_per_second": 0.29,
37
+ "step": 10
38
+ },
39
+ {
40
+ "epoch": 0.04,
41
+ "learning_rate": 9.80719794344473e-05,
42
+ "loss": 1.8616,
43
+ "step": 15
44
+ },
45
+ {
46
+ "epoch": 0.04,
47
+ "eval_loss": 1.8424055576324463,
48
+ "eval_runtime": 37.8727,
49
+ "eval_samples_per_second": 2.218,
50
+ "eval_steps_per_second": 0.29,
51
+ "step": 15
52
+ },
53
+ {
54
+ "epoch": 0.05,
55
+ "learning_rate": 9.742930591259641e-05,
56
+ "loss": 1.8267,
57
+ "step": 20
58
+ },
59
+ {
60
+ "epoch": 0.05,
61
+ "eval_loss": 1.8279685974121094,
62
+ "eval_runtime": 38.166,
63
+ "eval_samples_per_second": 2.201,
64
+ "eval_steps_per_second": 0.288,
65
+ "step": 20
66
+ },
67
+ {
68
+ "epoch": 0.06,
69
+ "learning_rate": 9.67866323907455e-05,
70
+ "loss": 1.835,
71
+ "step": 25
72
+ },
73
+ {
74
+ "epoch": 0.06,
75
+ "eval_loss": 1.8134318590164185,
76
+ "eval_runtime": 38.0411,
77
+ "eval_samples_per_second": 2.208,
78
+ "eval_steps_per_second": 0.289,
79
+ "step": 25
80
+ },
81
+ {
82
+ "epoch": 0.08,
83
+ "learning_rate": 9.61439588688946e-05,
84
+ "loss": 1.8236,
85
+ "step": 30
86
+ },
87
+ {
88
+ "epoch": 0.08,
89
+ "eval_loss": 1.8010995388031006,
90
+ "eval_runtime": 37.8753,
91
+ "eval_samples_per_second": 2.218,
92
+ "eval_steps_per_second": 0.29,
93
+ "step": 30
94
+ },
95
+ {
96
+ "epoch": 0.09,
97
+ "learning_rate": 9.550128534704372e-05,
98
+ "loss": 1.7186,
99
+ "step": 35
100
+ },
101
+ {
102
+ "epoch": 0.09,
103
+ "eval_loss": 1.7901933193206787,
104
+ "eval_runtime": 38.0374,
105
+ "eval_samples_per_second": 2.208,
106
+ "eval_steps_per_second": 0.289,
107
+ "step": 35
108
+ },
109
+ {
110
+ "epoch": 0.1,
111
+ "learning_rate": 9.485861182519281e-05,
112
+ "loss": 1.7959,
113
+ "step": 40
114
+ },
115
+ {
116
+ "epoch": 0.1,
117
+ "eval_loss": 1.7807564735412598,
118
+ "eval_runtime": 37.8784,
119
+ "eval_samples_per_second": 2.218,
120
+ "eval_steps_per_second": 0.29,
121
+ "step": 40
122
+ },
123
+ {
124
+ "epoch": 0.12,
125
+ "learning_rate": 9.421593830334192e-05,
126
+ "loss": 1.715,
127
+ "step": 45
128
+ },
129
+ {
130
+ "epoch": 0.12,
131
+ "eval_loss": 1.7709146738052368,
132
+ "eval_runtime": 37.8705,
133
+ "eval_samples_per_second": 2.218,
134
+ "eval_steps_per_second": 0.29,
135
+ "step": 45
136
+ },
137
+ {
138
+ "epoch": 0.13,
139
+ "learning_rate": 9.357326478149101e-05,
140
+ "loss": 1.7581,
141
+ "step": 50
142
+ },
143
+ {
144
+ "epoch": 0.13,
145
+ "eval_loss": 1.761339783668518,
146
+ "eval_runtime": 37.661,
147
+ "eval_samples_per_second": 2.23,
148
+ "eval_steps_per_second": 0.292,
149
+ "step": 50
150
+ },
151
+ {
152
+ "epoch": 0.14,
153
+ "learning_rate": 9.29305912596401e-05,
154
+ "loss": 1.7305,
155
+ "step": 55
156
+ },
157
+ {
158
+ "epoch": 0.14,
159
+ "eval_loss": 1.7389689683914185,
160
+ "eval_runtime": 38.0055,
161
+ "eval_samples_per_second": 2.21,
162
+ "eval_steps_per_second": 0.289,
163
+ "step": 55
164
+ },
165
+ {
166
+ "epoch": 0.15,
167
+ "learning_rate": 9.228791773778921e-05,
168
+ "loss": 1.7086,
169
+ "step": 60
170
+ },
171
+ {
172
+ "epoch": 0.15,
173
+ "eval_loss": 1.713218092918396,
174
+ "eval_runtime": 37.737,
175
+ "eval_samples_per_second": 2.226,
176
+ "eval_steps_per_second": 0.291,
177
+ "step": 60
178
+ },
179
+ {
180
+ "epoch": 0.17,
181
+ "learning_rate": 9.16452442159383e-05,
182
+ "loss": 1.7057,
183
+ "step": 65
184
+ },
185
+ {
186
+ "epoch": 0.17,
187
+ "eval_loss": 1.7108122110366821,
188
+ "eval_runtime": 38.0037,
189
+ "eval_samples_per_second": 2.21,
190
+ "eval_steps_per_second": 0.289,
191
+ "step": 65
192
+ },
193
+ {
194
+ "epoch": 0.18,
195
+ "learning_rate": 9.100257069408741e-05,
196
+ "loss": 1.7243,
197
+ "step": 70
198
+ },
199
+ {
200
+ "epoch": 0.18,
201
+ "eval_loss": 1.7041622400283813,
202
+ "eval_runtime": 38.1402,
203
+ "eval_samples_per_second": 2.202,
204
+ "eval_steps_per_second": 0.288,
205
+ "step": 70
206
+ },
207
+ {
208
+ "epoch": 0.19,
209
+ "learning_rate": 9.03598971722365e-05,
210
+ "loss": 1.7038,
211
+ "step": 75
212
+ },
213
+ {
214
+ "epoch": 0.19,
215
+ "eval_loss": 1.7010408639907837,
216
+ "eval_runtime": 37.7193,
217
+ "eval_samples_per_second": 2.227,
218
+ "eval_steps_per_second": 0.292,
219
+ "step": 75
220
+ },
221
+ {
222
+ "epoch": 0.21,
223
+ "learning_rate": 8.97172236503856e-05,
224
+ "loss": 1.6939,
225
+ "step": 80
226
+ },
227
+ {
228
+ "epoch": 0.21,
229
+ "eval_loss": 1.6987501382827759,
230
+ "eval_runtime": 38.0624,
231
+ "eval_samples_per_second": 2.207,
232
+ "eval_steps_per_second": 0.289,
233
+ "step": 80
234
+ },
235
+ {
236
+ "epoch": 0.22,
237
+ "learning_rate": 8.907455012853471e-05,
238
+ "loss": 1.645,
239
+ "step": 85
240
+ },
241
+ {
242
+ "epoch": 0.22,
243
+ "eval_loss": 1.6968218088150024,
244
+ "eval_runtime": 37.9205,
245
+ "eval_samples_per_second": 2.215,
246
+ "eval_steps_per_second": 0.29,
247
+ "step": 85
248
+ },
249
+ {
250
+ "epoch": 0.23,
251
+ "learning_rate": 8.84318766066838e-05,
252
+ "loss": 1.7016,
253
+ "step": 90
254
+ },
255
+ {
256
+ "epoch": 0.23,
257
+ "eval_loss": 1.6952500343322754,
258
+ "eval_runtime": 37.9617,
259
+ "eval_samples_per_second": 2.213,
260
+ "eval_steps_per_second": 0.29,
261
+ "step": 90
262
+ },
263
+ {
264
+ "epoch": 0.24,
265
+ "learning_rate": 8.778920308483291e-05,
266
+ "loss": 1.6869,
267
+ "step": 95
268
+ },
269
+ {
270
+ "epoch": 0.24,
271
+ "eval_loss": 1.6932854652404785,
272
+ "eval_runtime": 37.8781,
273
+ "eval_samples_per_second": 2.218,
274
+ "eval_steps_per_second": 0.29,
275
+ "step": 95
276
+ },
277
+ {
278
+ "epoch": 0.26,
279
+ "learning_rate": 8.7146529562982e-05,
280
+ "loss": 1.6728,
281
+ "step": 100
282
+ },
283
+ {
284
+ "epoch": 0.26,
285
+ "eval_loss": 1.6920738220214844,
286
+ "eval_runtime": 38.1996,
287
+ "eval_samples_per_second": 2.199,
288
+ "eval_steps_per_second": 0.288,
289
+ "step": 100
290
+ },
291
+ {
292
+ "epoch": 0.27,
293
+ "learning_rate": 8.650385604113111e-05,
294
+ "loss": 1.6674,
295
+ "step": 105
296
+ },
297
+ {
298
+ "epoch": 0.27,
299
+ "eval_loss": 1.6904469728469849,
300
+ "eval_runtime": 38.1252,
301
+ "eval_samples_per_second": 2.203,
302
+ "eval_steps_per_second": 0.289,
303
+ "step": 105
304
+ },
305
+ {
306
+ "epoch": 0.28,
307
+ "learning_rate": 8.586118251928022e-05,
308
+ "loss": 1.7013,
309
+ "step": 110
310
+ },
311
+ {
312
+ "epoch": 0.28,
313
+ "eval_loss": 1.6895124912261963,
314
+ "eval_runtime": 38.0968,
315
+ "eval_samples_per_second": 2.205,
316
+ "eval_steps_per_second": 0.289,
317
+ "step": 110
318
+ },
319
+ {
320
+ "epoch": 0.3,
321
+ "learning_rate": 8.521850899742931e-05,
322
+ "loss": 1.7148,
323
+ "step": 115
324
+ },
325
+ {
326
+ "epoch": 0.3,
327
+ "eval_loss": 1.6886374950408936,
328
+ "eval_runtime": 37.8799,
329
+ "eval_samples_per_second": 2.218,
330
+ "eval_steps_per_second": 0.29,
331
+ "step": 115
332
+ },
333
+ {
334
+ "epoch": 0.31,
335
+ "learning_rate": 8.457583547557842e-05,
336
+ "loss": 1.7166,
337
+ "step": 120
338
+ },
339
+ {
340
+ "epoch": 0.31,
341
+ "eval_loss": 1.6868164539337158,
342
+ "eval_runtime": 37.8861,
343
+ "eval_samples_per_second": 2.217,
344
+ "eval_steps_per_second": 0.29,
345
+ "step": 120
346
+ },
347
+ {
348
+ "epoch": 0.32,
349
+ "learning_rate": 8.393316195372751e-05,
350
+ "loss": 1.7012,
351
+ "step": 125
352
+ },
353
+ {
354
+ "epoch": 0.32,
355
+ "eval_loss": 1.6858367919921875,
356
+ "eval_runtime": 38.1136,
357
+ "eval_samples_per_second": 2.204,
358
+ "eval_steps_per_second": 0.289,
359
+ "step": 125
360
+ },
361
+ {
362
+ "epoch": 0.33,
363
+ "learning_rate": 8.32904884318766e-05,
364
+ "loss": 1.6827,
365
+ "step": 130
366
+ },
367
+ {
368
+ "epoch": 0.33,
369
+ "eval_loss": 1.6849361658096313,
370
+ "eval_runtime": 37.9883,
371
+ "eval_samples_per_second": 2.211,
372
+ "eval_steps_per_second": 0.29,
373
+ "step": 130
374
+ },
375
+ {
376
+ "epoch": 0.35,
377
+ "learning_rate": 8.264781491002571e-05,
378
+ "loss": 1.7157,
379
+ "step": 135
380
+ },
381
+ {
382
+ "epoch": 0.35,
383
+ "eval_loss": 1.684756875038147,
384
+ "eval_runtime": 37.8726,
385
+ "eval_samples_per_second": 2.218,
386
+ "eval_steps_per_second": 0.29,
387
+ "step": 135
388
+ },
389
+ {
390
+ "epoch": 0.36,
391
+ "learning_rate": 8.200514138817481e-05,
392
+ "loss": 1.6668,
393
+ "step": 140
394
+ },
395
+ {
396
+ "epoch": 0.36,
397
+ "eval_loss": 1.6836236715316772,
398
+ "eval_runtime": 37.9586,
399
+ "eval_samples_per_second": 2.213,
400
+ "eval_steps_per_second": 0.29,
401
+ "step": 140
402
+ },
403
+ {
404
+ "epoch": 0.37,
405
+ "learning_rate": 8.136246786632391e-05,
406
+ "loss": 1.6588,
407
+ "step": 145
408
+ },
409
+ {
410
+ "epoch": 0.37,
411
+ "eval_loss": 1.6828166246414185,
412
+ "eval_runtime": 38.0596,
413
+ "eval_samples_per_second": 2.207,
414
+ "eval_steps_per_second": 0.289,
415
+ "step": 145
416
+ },
417
+ {
418
+ "epoch": 0.39,
419
+ "learning_rate": 8.071979434447301e-05,
420
+ "loss": 1.7005,
421
+ "step": 150
422
+ },
423
+ {
424
+ "epoch": 0.39,
425
+ "eval_loss": 1.681463360786438,
426
+ "eval_runtime": 38.0878,
427
+ "eval_samples_per_second": 2.205,
428
+ "eval_steps_per_second": 0.289,
429
+ "step": 150
430
+ },
431
+ {
432
+ "epoch": 0.4,
433
+ "learning_rate": 8.007712082262212e-05,
434
+ "loss": 1.6893,
435
+ "step": 155
436
+ },
437
+ {
438
+ "epoch": 0.4,
439
+ "eval_loss": 1.6811552047729492,
440
+ "eval_runtime": 37.9265,
441
+ "eval_samples_per_second": 2.215,
442
+ "eval_steps_per_second": 0.29,
443
+ "step": 155
444
+ },
445
+ {
446
+ "epoch": 0.41,
447
+ "learning_rate": 7.943444730077121e-05,
448
+ "loss": 1.6682,
449
+ "step": 160
450
+ },
451
+ {
452
+ "epoch": 0.41,
453
+ "eval_loss": 1.6801096200942993,
454
+ "eval_runtime": 38.0375,
455
+ "eval_samples_per_second": 2.208,
456
+ "eval_steps_per_second": 0.289,
457
+ "step": 160
458
+ },
459
+ {
460
+ "epoch": 0.42,
461
+ "learning_rate": 7.87917737789203e-05,
462
+ "loss": 1.6481,
463
+ "step": 165
464
+ },
465
+ {
466
+ "epoch": 0.42,
467
+ "eval_loss": 1.6800144910812378,
468
+ "eval_runtime": 37.8916,
469
+ "eval_samples_per_second": 2.217,
470
+ "eval_steps_per_second": 0.29,
471
+ "step": 165
472
+ },
473
+ {
474
+ "epoch": 0.44,
475
+ "learning_rate": 7.814910025706941e-05,
476
+ "loss": 1.7042,
477
+ "step": 170
478
+ },
479
+ {
480
+ "epoch": 0.44,
481
+ "eval_loss": 1.6784976720809937,
482
+ "eval_runtime": 37.8514,
483
+ "eval_samples_per_second": 2.219,
484
+ "eval_steps_per_second": 0.291,
485
+ "step": 170
486
+ },
487
+ {
488
+ "epoch": 0.45,
489
+ "learning_rate": 7.750642673521852e-05,
490
+ "loss": 1.6555,
491
+ "step": 175
492
+ },
493
+ {
494
+ "epoch": 0.45,
495
+ "eval_loss": 1.6780468225479126,
496
+ "eval_runtime": 37.8893,
497
+ "eval_samples_per_second": 2.217,
498
+ "eval_steps_per_second": 0.29,
499
+ "step": 175
500
+ },
501
+ {
502
+ "epoch": 0.46,
503
+ "learning_rate": 7.686375321336761e-05,
504
+ "loss": 1.6717,
505
+ "step": 180
506
+ },
507
+ {
508
+ "epoch": 0.46,
509
+ "eval_loss": 1.6775085926055908,
510
+ "eval_runtime": 37.9289,
511
+ "eval_samples_per_second": 2.215,
512
+ "eval_steps_per_second": 0.29,
513
+ "step": 180
514
+ },
515
+ {
516
+ "epoch": 0.48,
517
+ "learning_rate": 7.622107969151672e-05,
518
+ "loss": 1.6716,
519
+ "step": 185
520
+ },
521
+ {
522
+ "epoch": 0.48,
523
+ "eval_loss": 1.6770914793014526,
524
+ "eval_runtime": 38.0111,
525
+ "eval_samples_per_second": 2.21,
526
+ "eval_steps_per_second": 0.289,
527
+ "step": 185
528
+ },
529
+ {
530
+ "epoch": 0.49,
531
+ "learning_rate": 7.557840616966581e-05,
532
+ "loss": 1.6618,
533
+ "step": 190
534
+ },
535
+ {
536
+ "epoch": 0.49,
537
+ "eval_loss": 1.676563024520874,
538
+ "eval_runtime": 38.0703,
539
+ "eval_samples_per_second": 2.206,
540
+ "eval_steps_per_second": 0.289,
541
+ "step": 190
542
+ },
543
+ {
544
+ "epoch": 0.5,
545
+ "learning_rate": 7.493573264781492e-05,
546
+ "loss": 1.6411,
547
+ "step": 195
548
+ },
549
+ {
550
+ "epoch": 0.5,
551
+ "eval_loss": 1.6757832765579224,
552
+ "eval_runtime": 37.827,
553
+ "eval_samples_per_second": 2.221,
554
+ "eval_steps_per_second": 0.291,
555
+ "step": 195
556
+ },
557
+ {
558
+ "epoch": 0.51,
559
+ "learning_rate": 7.429305912596401e-05,
560
+ "loss": 1.6637,
561
+ "step": 200
562
+ },
563
+ {
564
+ "epoch": 0.51,
565
+ "eval_loss": 1.6751985549926758,
566
+ "eval_runtime": 37.836,
567
+ "eval_samples_per_second": 2.22,
568
+ "eval_steps_per_second": 0.291,
569
+ "step": 200
570
+ },
571
+ {
572
+ "epoch": 0.53,
573
+ "learning_rate": 7.365038560411311e-05,
574
+ "loss": 1.6142,
575
+ "step": 205
576
+ },
577
+ {
578
+ "epoch": 0.53,
579
+ "eval_loss": 1.675147294998169,
580
+ "eval_runtime": 37.9175,
581
+ "eval_samples_per_second": 2.215,
582
+ "eval_steps_per_second": 0.29,
583
+ "step": 205
584
+ },
585
+ {
586
+ "epoch": 0.54,
587
+ "learning_rate": 7.300771208226222e-05,
588
+ "loss": 1.6741,
589
+ "step": 210
590
+ },
591
+ {
592
+ "epoch": 0.54,
593
+ "eval_loss": 1.674437165260315,
594
+ "eval_runtime": 38.087,
595
+ "eval_samples_per_second": 2.205,
596
+ "eval_steps_per_second": 0.289,
597
+ "step": 210
598
+ },
599
+ {
600
+ "epoch": 0.55,
601
+ "learning_rate": 7.236503856041131e-05,
602
+ "loss": 1.6408,
603
+ "step": 215
604
+ },
605
+ {
606
+ "epoch": 0.55,
607
+ "eval_loss": 1.6737719774246216,
608
+ "eval_runtime": 38.0075,
609
+ "eval_samples_per_second": 2.21,
610
+ "eval_steps_per_second": 0.289,
611
+ "step": 215
612
+ },
613
+ {
614
+ "epoch": 0.57,
615
+ "learning_rate": 7.172236503856042e-05,
616
+ "loss": 1.6733,
617
+ "step": 220
618
+ },
619
+ {
620
+ "epoch": 0.57,
621
+ "eval_loss": 1.6732759475708008,
622
+ "eval_runtime": 38.1125,
623
+ "eval_samples_per_second": 2.204,
624
+ "eval_steps_per_second": 0.289,
625
+ "step": 220
626
+ },
627
+ {
628
+ "epoch": 0.58,
629
+ "learning_rate": 7.107969151670951e-05,
630
+ "loss": 1.679,
631
+ "step": 225
632
+ },
633
+ {
634
+ "epoch": 0.58,
635
+ "eval_loss": 1.6726195812225342,
636
+ "eval_runtime": 37.9769,
637
+ "eval_samples_per_second": 2.212,
638
+ "eval_steps_per_second": 0.29,
639
+ "step": 225
640
+ },
641
+ {
642
+ "epoch": 0.59,
643
+ "learning_rate": 7.043701799485862e-05,
644
+ "loss": 1.7202,
645
+ "step": 230
646
+ },
647
+ {
648
+ "epoch": 0.59,
649
+ "eval_loss": 1.671908974647522,
650
+ "eval_runtime": 37.8215,
651
+ "eval_samples_per_second": 2.221,
652
+ "eval_steps_per_second": 0.291,
653
+ "step": 230
654
+ },
655
+ {
656
+ "epoch": 0.6,
657
+ "learning_rate": 6.979434447300771e-05,
658
+ "loss": 1.6805,
659
+ "step": 235
660
+ },
661
+ {
662
+ "epoch": 0.6,
663
+ "eval_loss": 1.6715577840805054,
664
+ "eval_runtime": 37.9343,
665
+ "eval_samples_per_second": 2.214,
666
+ "eval_steps_per_second": 0.29,
667
+ "step": 235
668
+ },
669
+ {
670
+ "epoch": 0.62,
671
+ "learning_rate": 6.91516709511568e-05,
672
+ "loss": 1.6331,
673
+ "step": 240
674
+ },
675
+ {
676
+ "epoch": 0.62,
677
+ "eval_loss": 1.6715376377105713,
678
+ "eval_runtime": 37.9904,
679
+ "eval_samples_per_second": 2.211,
680
+ "eval_steps_per_second": 0.29,
681
+ "step": 240
682
+ },
683
+ {
684
+ "epoch": 0.63,
685
+ "learning_rate": 6.850899742930593e-05,
686
+ "loss": 1.6761,
687
+ "step": 245
688
+ },
689
+ {
690
+ "epoch": 0.63,
691
+ "eval_loss": 1.671446681022644,
692
+ "eval_runtime": 38.0357,
693
+ "eval_samples_per_second": 2.208,
694
+ "eval_steps_per_second": 0.289,
695
+ "step": 245
696
+ },
697
+ {
698
+ "epoch": 0.64,
699
+ "learning_rate": 6.786632390745502e-05,
700
+ "loss": 1.6994,
701
+ "step": 250
702
+ },
703
+ {
704
+ "epoch": 0.64,
705
+ "eval_loss": 1.6716604232788086,
706
+ "eval_runtime": 37.9609,
707
+ "eval_samples_per_second": 2.213,
708
+ "eval_steps_per_second": 0.29,
709
+ "step": 250
710
+ },
711
+ {
712
+ "epoch": 0.66,
713
+ "learning_rate": 6.722365038560411e-05,
714
+ "loss": 1.6305,
715
+ "step": 255
716
+ },
717
+ {
718
+ "epoch": 0.66,
719
+ "eval_loss": 1.6711723804473877,
720
+ "eval_runtime": 37.7893,
721
+ "eval_samples_per_second": 2.223,
722
+ "eval_steps_per_second": 0.291,
723
+ "step": 255
724
+ },
725
+ {
726
+ "epoch": 0.67,
727
+ "learning_rate": 6.658097686375322e-05,
728
+ "loss": 1.6612,
729
+ "step": 260
730
+ },
731
+ {
732
+ "epoch": 0.67,
733
+ "eval_loss": 1.670398473739624,
734
+ "eval_runtime": 37.7667,
735
+ "eval_samples_per_second": 2.224,
736
+ "eval_steps_per_second": 0.291,
737
+ "step": 260
738
+ },
739
+ {
740
+ "epoch": 0.68,
741
+ "learning_rate": 6.593830334190231e-05,
742
+ "loss": 1.6576,
743
+ "step": 265
744
+ },
745
+ {
746
+ "epoch": 0.68,
747
+ "eval_loss": 1.6706459522247314,
748
+ "eval_runtime": 37.8285,
749
+ "eval_samples_per_second": 2.221,
750
+ "eval_steps_per_second": 0.291,
751
+ "step": 265
752
+ },
753
+ {
754
+ "epoch": 0.69,
755
+ "learning_rate": 6.529562982005142e-05,
756
+ "loss": 1.6837,
757
+ "step": 270
758
+ },
759
+ {
760
+ "epoch": 0.69,
761
+ "eval_loss": 1.6699285507202148,
762
+ "eval_runtime": 37.7619,
763
+ "eval_samples_per_second": 2.224,
764
+ "eval_steps_per_second": 0.291,
765
+ "step": 270
766
+ },
767
+ {
768
+ "epoch": 0.71,
769
+ "learning_rate": 6.465295629820052e-05,
770
+ "loss": 1.6493,
771
+ "step": 275
772
+ },
773
+ {
774
+ "epoch": 0.71,
775
+ "eval_loss": 1.6693135499954224,
776
+ "eval_runtime": 37.7441,
777
+ "eval_samples_per_second": 2.226,
778
+ "eval_steps_per_second": 0.291,
779
+ "step": 275
780
+ },
781
+ {
782
+ "epoch": 0.72,
783
+ "learning_rate": 6.401028277634962e-05,
784
+ "loss": 1.6427,
785
+ "step": 280
786
+ },
787
+ {
788
+ "epoch": 0.72,
789
+ "eval_loss": 1.669467568397522,
790
+ "eval_runtime": 37.8707,
791
+ "eval_samples_per_second": 2.218,
792
+ "eval_steps_per_second": 0.29,
793
+ "step": 280
794
+ },
795
+ {
796
+ "epoch": 0.73,
797
+ "learning_rate": 6.336760925449872e-05,
798
+ "loss": 1.6236,
799
+ "step": 285
800
+ },
801
+ {
802
+ "epoch": 0.73,
803
+ "eval_loss": 1.6690431833267212,
804
+ "eval_runtime": 37.8946,
805
+ "eval_samples_per_second": 2.217,
806
+ "eval_steps_per_second": 0.29,
807
+ "step": 285
808
+ },
809
+ {
810
+ "epoch": 0.75,
811
+ "learning_rate": 6.272493573264781e-05,
812
+ "loss": 1.6538,
813
+ "step": 290
814
+ },
815
+ {
816
+ "epoch": 0.75,
817
+ "eval_loss": 1.6688117980957031,
818
+ "eval_runtime": 37.9378,
819
+ "eval_samples_per_second": 2.214,
820
+ "eval_steps_per_second": 0.29,
821
+ "step": 290
822
+ },
823
+ {
824
+ "epoch": 0.76,
825
+ "learning_rate": 6.208226221079692e-05,
826
+ "loss": 1.6558,
827
+ "step": 295
828
+ },
829
+ {
830
+ "epoch": 0.76,
831
+ "eval_loss": 1.668560266494751,
832
+ "eval_runtime": 38.0448,
833
+ "eval_samples_per_second": 2.208,
834
+ "eval_steps_per_second": 0.289,
835
+ "step": 295
836
+ },
837
+ {
838
+ "epoch": 0.77,
839
+ "learning_rate": 6.143958868894601e-05,
840
+ "loss": 1.6401,
841
+ "step": 300
842
+ },
843
+ {
844
+ "epoch": 0.77,
845
+ "eval_loss": 1.6680612564086914,
846
+ "eval_runtime": 37.8571,
847
+ "eval_samples_per_second": 2.219,
848
+ "eval_steps_per_second": 0.291,
849
+ "step": 300
850
+ },
851
+ {
852
+ "epoch": 0.78,
853
+ "learning_rate": 6.079691516709511e-05,
854
+ "loss": 1.6468,
855
+ "step": 305
856
+ },
857
+ {
858
+ "epoch": 0.78,
859
+ "eval_loss": 1.6675914525985718,
860
+ "eval_runtime": 37.8491,
861
+ "eval_samples_per_second": 2.219,
862
+ "eval_steps_per_second": 0.291,
863
+ "step": 305
864
+ },
865
+ {
866
+ "epoch": 0.8,
867
+ "learning_rate": 6.015424164524421e-05,
868
+ "loss": 1.6579,
869
+ "step": 310
870
+ },
871
+ {
872
+ "epoch": 0.8,
873
+ "eval_loss": 1.6671632528305054,
874
+ "eval_runtime": 37.7658,
875
+ "eval_samples_per_second": 2.224,
876
+ "eval_steps_per_second": 0.291,
877
+ "step": 310
878
+ },
879
+ {
880
+ "epoch": 0.81,
881
+ "learning_rate": 5.951156812339333e-05,
882
+ "loss": 1.6339,
883
+ "step": 315
884
+ },
885
+ {
886
+ "epoch": 0.81,
887
+ "eval_loss": 1.666803240776062,
888
+ "eval_runtime": 37.7896,
889
+ "eval_samples_per_second": 2.223,
890
+ "eval_steps_per_second": 0.291,
891
+ "step": 315
892
+ },
893
+ {
894
+ "epoch": 0.82,
895
+ "learning_rate": 5.886889460154242e-05,
896
+ "loss": 1.6636,
897
+ "step": 320
898
+ },
899
+ {
900
+ "epoch": 0.82,
901
+ "eval_loss": 1.6664105653762817,
902
+ "eval_runtime": 38.0586,
903
+ "eval_samples_per_second": 2.207,
904
+ "eval_steps_per_second": 0.289,
905
+ "step": 320
906
+ },
907
+ {
908
+ "epoch": 0.84,
909
+ "learning_rate": 5.822622107969152e-05,
910
+ "loss": 1.641,
911
+ "step": 325
912
+ },
913
+ {
914
+ "epoch": 0.84,
915
+ "eval_loss": 1.6665875911712646,
916
+ "eval_runtime": 37.9046,
917
+ "eval_samples_per_second": 2.216,
918
+ "eval_steps_per_second": 0.29,
919
+ "step": 325
920
+ },
921
+ {
922
+ "epoch": 0.85,
923
+ "learning_rate": 5.758354755784062e-05,
924
+ "loss": 1.6616,
925
+ "step": 330
926
+ },
927
+ {
928
+ "epoch": 0.85,
929
+ "eval_loss": 1.6664165258407593,
930
+ "eval_runtime": 37.8575,
931
+ "eval_samples_per_second": 2.219,
932
+ "eval_steps_per_second": 0.291,
933
+ "step": 330
934
+ },
935
+ {
936
+ "epoch": 0.86,
937
+ "learning_rate": 5.694087403598972e-05,
938
+ "loss": 1.6597,
939
+ "step": 335
940
+ },
941
+ {
942
+ "epoch": 0.86,
943
+ "eval_loss": 1.666045904159546,
944
+ "eval_runtime": 37.9711,
945
+ "eval_samples_per_second": 2.212,
946
+ "eval_steps_per_second": 0.29,
947
+ "step": 335
948
+ },
949
+ {
950
+ "epoch": 0.87,
951
+ "learning_rate": 5.6298200514138824e-05,
952
+ "loss": 1.67,
953
+ "step": 340
954
+ },
955
+ {
956
+ "epoch": 0.87,
957
+ "eval_loss": 1.666337490081787,
958
+ "eval_runtime": 37.8821,
959
+ "eval_samples_per_second": 2.217,
960
+ "eval_steps_per_second": 0.29,
961
+ "step": 340
962
+ },
963
+ {
964
+ "epoch": 0.89,
965
+ "learning_rate": 5.5655526992287924e-05,
966
+ "loss": 1.6344,
967
+ "step": 345
968
+ },
969
+ {
970
+ "epoch": 0.89,
971
+ "eval_loss": 1.6659029722213745,
972
+ "eval_runtime": 37.9188,
973
+ "eval_samples_per_second": 2.215,
974
+ "eval_steps_per_second": 0.29,
975
+ "step": 345
976
+ },
977
+ {
978
+ "epoch": 0.9,
979
+ "learning_rate": 5.501285347043702e-05,
980
+ "loss": 1.6623,
981
+ "step": 350
982
+ },
983
+ {
984
+ "epoch": 0.9,
985
+ "eval_loss": 1.6656525135040283,
986
+ "eval_runtime": 37.9951,
987
+ "eval_samples_per_second": 2.211,
988
+ "eval_steps_per_second": 0.29,
989
+ "step": 350
990
+ },
991
+ {
992
+ "epoch": 0.91,
993
+ "learning_rate": 5.437017994858612e-05,
994
+ "loss": 1.6623,
995
+ "step": 355
996
+ },
997
+ {
998
+ "epoch": 0.91,
999
+ "eval_loss": 1.6654421091079712,
1000
+ "eval_runtime": 37.8803,
1001
+ "eval_samples_per_second": 2.218,
1002
+ "eval_steps_per_second": 0.29,
1003
+ "step": 355
1004
+ },
1005
+ {
1006
+ "epoch": 0.93,
1007
+ "learning_rate": 5.372750642673522e-05,
1008
+ "loss": 1.6741,
1009
+ "step": 360
1010
+ },
1011
+ {
1012
+ "epoch": 0.93,
1013
+ "eval_loss": 1.665541648864746,
1014
+ "eval_runtime": 37.7151,
1015
+ "eval_samples_per_second": 2.227,
1016
+ "eval_steps_per_second": 0.292,
1017
+ "step": 360
1018
+ },
1019
+ {
1020
+ "epoch": 0.94,
1021
+ "learning_rate": 5.308483290488432e-05,
1022
+ "loss": 1.6973,
1023
+ "step": 365
1024
+ },
1025
+ {
1026
+ "epoch": 0.94,
1027
+ "eval_loss": 1.6651471853256226,
1028
+ "eval_runtime": 37.9144,
1029
+ "eval_samples_per_second": 2.216,
1030
+ "eval_steps_per_second": 0.29,
1031
+ "step": 365
1032
+ },
1033
+ {
1034
+ "epoch": 0.95,
1035
+ "learning_rate": 5.244215938303342e-05,
1036
+ "loss": 1.6829,
1037
+ "step": 370
1038
+ },
1039
+ {
1040
+ "epoch": 0.95,
1041
+ "eval_loss": 1.6651064157485962,
1042
+ "eval_runtime": 37.7443,
1043
+ "eval_samples_per_second": 2.226,
1044
+ "eval_steps_per_second": 0.291,
1045
+ "step": 370
1046
+ },
1047
+ {
1048
+ "epoch": 0.96,
1049
+ "learning_rate": 5.1799485861182514e-05,
1050
+ "loss": 1.6923,
1051
+ "step": 375
1052
+ },
1053
+ {
1054
+ "epoch": 0.96,
1055
+ "eval_loss": 1.6646034717559814,
1056
+ "eval_runtime": 37.8783,
1057
+ "eval_samples_per_second": 2.218,
1058
+ "eval_steps_per_second": 0.29,
1059
+ "step": 375
1060
+ },
1061
+ {
1062
+ "epoch": 0.98,
1063
+ "learning_rate": 5.1156812339331615e-05,
1064
+ "loss": 1.6725,
1065
+ "step": 380
1066
+ },
1067
+ {
1068
+ "epoch": 0.98,
1069
+ "eval_loss": 1.6639235019683838,
1070
+ "eval_runtime": 37.7559,
1071
+ "eval_samples_per_second": 2.225,
1072
+ "eval_steps_per_second": 0.291,
1073
+ "step": 380
1074
+ },
1075
+ {
1076
+ "epoch": 0.99,
1077
+ "learning_rate": 5.051413881748073e-05,
1078
+ "loss": 1.6216,
1079
+ "step": 385
1080
+ },
1081
+ {
1082
+ "epoch": 0.99,
1083
+ "eval_loss": 1.6643506288528442,
1084
+ "eval_runtime": 37.9895,
1085
+ "eval_samples_per_second": 2.211,
1086
+ "eval_steps_per_second": 0.29,
1087
+ "step": 385
1088
+ },
1089
+ {
1090
+ "epoch": 1.0,
1091
+ "learning_rate": 4.987146529562982e-05,
1092
+ "loss": 1.6518,
1093
+ "step": 390
1094
+ },
1095
+ {
1096
+ "epoch": 1.0,
1097
+ "eval_loss": 1.6643046140670776,
1098
+ "eval_runtime": 37.9289,
1099
+ "eval_samples_per_second": 2.215,
1100
+ "eval_steps_per_second": 0.29,
1101
+ "step": 390
1102
+ },
1103
+ {
1104
+ "epoch": 1.02,
1105
+ "learning_rate": 4.922879177377892e-05,
1106
+ "loss": 1.658,
1107
+ "step": 395
1108
+ },
1109
+ {
1110
+ "epoch": 1.02,
1111
+ "eval_loss": 1.6641957759857178,
1112
+ "eval_runtime": 37.8638,
1113
+ "eval_samples_per_second": 2.218,
1114
+ "eval_steps_per_second": 0.291,
1115
+ "step": 395
1116
+ },
1117
+ {
1118
+ "epoch": 1.03,
1119
+ "learning_rate": 4.8586118251928024e-05,
1120
+ "loss": 1.6767,
1121
+ "step": 400
1122
+ },
1123
+ {
1124
+ "epoch": 1.03,
1125
+ "eval_loss": 1.663757085800171,
1126
+ "eval_runtime": 37.7728,
1127
+ "eval_samples_per_second": 2.224,
1128
+ "eval_steps_per_second": 0.291,
1129
+ "step": 400
1130
+ },
1131
+ {
1132
+ "epoch": 1.04,
1133
+ "learning_rate": 4.7943444730077124e-05,
1134
+ "loss": 1.6264,
1135
+ "step": 405
1136
+ },
1137
+ {
1138
+ "epoch": 1.04,
1139
+ "eval_loss": 1.6635041236877441,
1140
+ "eval_runtime": 37.7702,
1141
+ "eval_samples_per_second": 2.224,
1142
+ "eval_steps_per_second": 0.291,
1143
+ "step": 405
1144
+ },
1145
+ {
1146
+ "epoch": 1.05,
1147
+ "learning_rate": 4.7300771208226225e-05,
1148
+ "loss": 1.6527,
1149
+ "step": 410
1150
+ },
1151
+ {
1152
+ "epoch": 1.05,
1153
+ "eval_loss": 1.6632280349731445,
1154
+ "eval_runtime": 37.8453,
1155
+ "eval_samples_per_second": 2.22,
1156
+ "eval_steps_per_second": 0.291,
1157
+ "step": 410
1158
+ },
1159
+ {
1160
+ "epoch": 1.07,
1161
+ "learning_rate": 4.6658097686375325e-05,
1162
+ "loss": 1.6157,
1163
+ "step": 415
1164
+ },
1165
+ {
1166
+ "epoch": 1.07,
1167
+ "eval_loss": 1.6636598110198975,
1168
+ "eval_runtime": 37.812,
1169
+ "eval_samples_per_second": 2.222,
1170
+ "eval_steps_per_second": 0.291,
1171
+ "step": 415
1172
+ },
1173
+ {
1174
+ "epoch": 1.08,
1175
+ "learning_rate": 4.6015424164524426e-05,
1176
+ "loss": 1.5966,
1177
+ "step": 420
1178
+ },
1179
+ {
1180
+ "epoch": 1.08,
1181
+ "eval_loss": 1.663393497467041,
1182
+ "eval_runtime": 37.7743,
1183
+ "eval_samples_per_second": 2.224,
1184
+ "eval_steps_per_second": 0.291,
1185
+ "step": 420
1186
+ },
1187
+ {
1188
+ "epoch": 1.09,
1189
+ "learning_rate": 4.537275064267352e-05,
1190
+ "loss": 1.6705,
1191
+ "step": 425
1192
+ },
1193
+ {
1194
+ "epoch": 1.09,
1195
+ "eval_loss": 1.6631269454956055,
1196
+ "eval_runtime": 38.0953,
1197
+ "eval_samples_per_second": 2.205,
1198
+ "eval_steps_per_second": 0.289,
1199
+ "step": 425
1200
+ },
1201
+ {
1202
+ "epoch": 1.11,
1203
+ "learning_rate": 4.473007712082262e-05,
1204
+ "loss": 1.6691,
1205
+ "step": 430
1206
+ },
1207
+ {
1208
+ "epoch": 1.11,
1209
+ "eval_loss": 1.6633110046386719,
1210
+ "eval_runtime": 37.8964,
1211
+ "eval_samples_per_second": 2.217,
1212
+ "eval_steps_per_second": 0.29,
1213
+ "step": 430
1214
+ },
1215
+ {
1216
+ "epoch": 1.12,
1217
+ "learning_rate": 4.408740359897173e-05,
1218
+ "loss": 1.6332,
1219
+ "step": 435
1220
+ },
1221
+ {
1222
+ "epoch": 1.12,
1223
+ "eval_loss": 1.6628649234771729,
1224
+ "eval_runtime": 38.0621,
1225
+ "eval_samples_per_second": 2.207,
1226
+ "eval_steps_per_second": 0.289,
1227
+ "step": 435
1228
+ },
1229
+ {
1230
+ "epoch": 1.13,
1231
+ "learning_rate": 4.344473007712083e-05,
1232
+ "loss": 1.5916,
1233
+ "step": 440
1234
+ },
1235
+ {
1236
+ "epoch": 1.13,
1237
+ "eval_loss": 1.662834882736206,
1238
+ "eval_runtime": 37.9722,
1239
+ "eval_samples_per_second": 2.212,
1240
+ "eval_steps_per_second": 0.29,
1241
+ "step": 440
1242
+ },
1243
+ {
1244
+ "epoch": 1.14,
1245
+ "learning_rate": 4.280205655526993e-05,
1246
+ "loss": 1.6543,
1247
+ "step": 445
1248
+ },
1249
+ {
1250
+ "epoch": 1.14,
1251
+ "eval_loss": 1.6630265712738037,
1252
+ "eval_runtime": 37.9699,
1253
+ "eval_samples_per_second": 2.212,
1254
+ "eval_steps_per_second": 0.29,
1255
+ "step": 445
1256
+ },
1257
+ {
1258
+ "epoch": 1.16,
1259
+ "learning_rate": 4.215938303341902e-05,
1260
+ "loss": 1.6353,
1261
+ "step": 450
1262
+ },
1263
+ {
1264
+ "epoch": 1.16,
1265
+ "eval_loss": 1.6625033617019653,
1266
+ "eval_runtime": 37.7505,
1267
+ "eval_samples_per_second": 2.225,
1268
+ "eval_steps_per_second": 0.291,
1269
+ "step": 450
1270
+ },
1271
+ {
1272
+ "epoch": 1.17,
1273
+ "learning_rate": 4.151670951156812e-05,
1274
+ "loss": 1.6441,
1275
+ "step": 455
1276
+ },
1277
+ {
1278
+ "epoch": 1.17,
1279
+ "eval_loss": 1.662785530090332,
1280
+ "eval_runtime": 37.75,
1281
+ "eval_samples_per_second": 2.225,
1282
+ "eval_steps_per_second": 0.291,
1283
+ "step": 455
1284
+ },
1285
+ {
1286
+ "epoch": 1.18,
1287
+ "learning_rate": 4.0874035989717224e-05,
1288
+ "loss": 1.6631,
1289
+ "step": 460
1290
+ },
1291
+ {
1292
+ "epoch": 1.18,
1293
+ "eval_loss": 1.6627779006958008,
1294
+ "eval_runtime": 37.864,
1295
+ "eval_samples_per_second": 2.218,
1296
+ "eval_steps_per_second": 0.291,
1297
+ "step": 460
1298
+ },
1299
+ {
1300
+ "epoch": 1.2,
1301
+ "learning_rate": 4.0231362467866324e-05,
1302
+ "loss": 1.6327,
1303
+ "step": 465
1304
+ },
1305
+ {
1306
+ "epoch": 1.2,
1307
+ "eval_loss": 1.662458896636963,
1308
+ "eval_runtime": 37.7496,
1309
+ "eval_samples_per_second": 2.225,
1310
+ "eval_steps_per_second": 0.291,
1311
+ "step": 465
1312
+ },
1313
+ {
1314
+ "epoch": 1.21,
1315
+ "learning_rate": 3.958868894601543e-05,
1316
+ "loss": 1.6238,
1317
+ "step": 470
1318
+ },
1319
+ {
1320
+ "epoch": 1.21,
1321
+ "eval_loss": 1.6621876955032349,
1322
+ "eval_runtime": 37.8943,
1323
+ "eval_samples_per_second": 2.217,
1324
+ "eval_steps_per_second": 0.29,
1325
+ "step": 470
1326
+ },
1327
+ {
1328
+ "epoch": 1.22,
1329
+ "learning_rate": 3.8946015424164526e-05,
1330
+ "loss": 1.6231,
1331
+ "step": 475
1332
+ },
1333
+ {
1334
+ "epoch": 1.22,
1335
+ "eval_loss": 1.662013292312622,
1336
+ "eval_runtime": 37.9364,
1337
+ "eval_samples_per_second": 2.214,
1338
+ "eval_steps_per_second": 0.29,
1339
+ "step": 475
1340
+ },
1341
+ {
1342
+ "epoch": 1.23,
1343
+ "learning_rate": 3.8303341902313626e-05,
1344
+ "loss": 1.6381,
1345
+ "step": 480
1346
+ },
1347
+ {
1348
+ "epoch": 1.23,
1349
+ "eval_loss": 1.6615785360336304,
1350
+ "eval_runtime": 37.91,
1351
+ "eval_samples_per_second": 2.216,
1352
+ "eval_steps_per_second": 0.29,
1353
+ "step": 480
1354
+ },
1355
+ {
1356
+ "epoch": 1.25,
1357
+ "learning_rate": 3.766066838046273e-05,
1358
+ "loss": 1.6419,
1359
+ "step": 485
1360
+ },
1361
+ {
1362
+ "epoch": 1.25,
1363
+ "eval_loss": 1.662042498588562,
1364
+ "eval_runtime": 37.7834,
1365
+ "eval_samples_per_second": 2.223,
1366
+ "eval_steps_per_second": 0.291,
1367
+ "step": 485
1368
+ },
1369
+ {
1370
+ "epoch": 1.26,
1371
+ "learning_rate": 3.701799485861183e-05,
1372
+ "loss": 1.6319,
1373
+ "step": 490
1374
+ },
1375
+ {
1376
+ "epoch": 1.26,
1377
+ "eval_loss": 1.6619195938110352,
1378
+ "eval_runtime": 37.7385,
1379
+ "eval_samples_per_second": 2.226,
1380
+ "eval_steps_per_second": 0.291,
1381
+ "step": 490
1382
+ },
1383
+ {
1384
+ "epoch": 1.27,
1385
+ "learning_rate": 3.637532133676093e-05,
1386
+ "loss": 1.609,
1387
+ "step": 495
1388
+ },
1389
+ {
1390
+ "epoch": 1.27,
1391
+ "eval_loss": 1.6615225076675415,
1392
+ "eval_runtime": 37.8056,
1393
+ "eval_samples_per_second": 2.222,
1394
+ "eval_steps_per_second": 0.291,
1395
+ "step": 495
1396
+ },
1397
+ {
1398
+ "epoch": 1.29,
1399
+ "learning_rate": 3.573264781491003e-05,
1400
+ "loss": 1.6435,
1401
+ "step": 500
1402
+ },
1403
+ {
1404
+ "epoch": 1.29,
1405
+ "eval_loss": 1.6610548496246338,
1406
+ "eval_runtime": 37.9353,
1407
+ "eval_samples_per_second": 2.214,
1408
+ "eval_steps_per_second": 0.29,
1409
+ "step": 500
1410
+ },
1411
+ {
1412
+ "epoch": 1.3,
1413
+ "learning_rate": 3.508997429305913e-05,
1414
+ "loss": 1.6381,
1415
+ "step": 505
1416
+ },
1417
+ {
1418
+ "epoch": 1.3,
1419
+ "eval_loss": 1.661049246788025,
1420
+ "eval_runtime": 37.9669,
1421
+ "eval_samples_per_second": 2.212,
1422
+ "eval_steps_per_second": 0.29,
1423
+ "step": 505
1424
+ },
1425
+ {
1426
+ "epoch": 1.31,
1427
+ "learning_rate": 3.444730077120823e-05,
1428
+ "loss": 1.6506,
1429
+ "step": 510
1430
+ },
1431
+ {
1432
+ "epoch": 1.31,
1433
+ "eval_loss": 1.6610089540481567,
1434
+ "eval_runtime": 37.9127,
1435
+ "eval_samples_per_second": 2.216,
1436
+ "eval_steps_per_second": 0.29,
1437
+ "step": 510
1438
+ },
1439
+ {
1440
+ "epoch": 1.32,
1441
+ "learning_rate": 3.380462724935733e-05,
1442
+ "loss": 1.6376,
1443
+ "step": 515
1444
+ },
1445
+ {
1446
+ "epoch": 1.32,
1447
+ "eval_loss": 1.6608952283859253,
1448
+ "eval_runtime": 37.9454,
1449
+ "eval_samples_per_second": 2.214,
1450
+ "eval_steps_per_second": 0.29,
1451
+ "step": 515
1452
+ },
1453
+ {
1454
+ "epoch": 1.34,
1455
+ "learning_rate": 3.316195372750643e-05,
1456
+ "loss": 1.6354,
1457
+ "step": 520
1458
+ },
1459
+ {
1460
+ "epoch": 1.34,
1461
+ "eval_loss": 1.6610838174819946,
1462
+ "eval_runtime": 37.7265,
1463
+ "eval_samples_per_second": 2.227,
1464
+ "eval_steps_per_second": 0.292,
1465
+ "step": 520
1466
+ },
1467
+ {
1468
+ "epoch": 1.35,
1469
+ "learning_rate": 3.251928020565553e-05,
1470
+ "loss": 1.6201,
1471
+ "step": 525
1472
+ },
1473
+ {
1474
+ "epoch": 1.35,
1475
+ "eval_loss": 1.6610314846038818,
1476
+ "eval_runtime": 37.8464,
1477
+ "eval_samples_per_second": 2.219,
1478
+ "eval_steps_per_second": 0.291,
1479
+ "step": 525
1480
+ },
1481
+ {
1482
+ "epoch": 1.36,
1483
+ "learning_rate": 3.1876606683804625e-05,
1484
+ "loss": 1.6461,
1485
+ "step": 530
1486
+ },
1487
+ {
1488
+ "epoch": 1.36,
1489
+ "eval_loss": 1.6605802774429321,
1490
+ "eval_runtime": 37.9503,
1491
+ "eval_samples_per_second": 2.213,
1492
+ "eval_steps_per_second": 0.29,
1493
+ "step": 530
1494
+ },
1495
+ {
1496
+ "epoch": 1.38,
1497
+ "learning_rate": 3.1233933161953726e-05,
1498
+ "loss": 1.6818,
1499
+ "step": 535
1500
+ },
1501
+ {
1502
+ "epoch": 1.38,
1503
+ "eval_loss": 1.6607571840286255,
1504
+ "eval_runtime": 37.9086,
1505
+ "eval_samples_per_second": 2.216,
1506
+ "eval_steps_per_second": 0.29,
1507
+ "step": 535
1508
+ },
1509
+ {
1510
+ "epoch": 1.39,
1511
+ "learning_rate": 3.059125964010283e-05,
1512
+ "loss": 1.6117,
1513
+ "step": 540
1514
+ },
1515
+ {
1516
+ "epoch": 1.39,
1517
+ "eval_loss": 1.6605336666107178,
1518
+ "eval_runtime": 37.947,
1519
+ "eval_samples_per_second": 2.214,
1520
+ "eval_steps_per_second": 0.29,
1521
+ "step": 540
1522
+ },
1523
+ {
1524
+ "epoch": 1.4,
1525
+ "learning_rate": 2.994858611825193e-05,
1526
+ "loss": 1.6252,
1527
+ "step": 545
1528
+ },
1529
+ {
1530
+ "epoch": 1.4,
1531
+ "eval_loss": 1.6603213548660278,
1532
+ "eval_runtime": 37.9826,
1533
+ "eval_samples_per_second": 2.212,
1534
+ "eval_steps_per_second": 0.29,
1535
+ "step": 545
1536
+ },
1537
+ {
1538
+ "epoch": 1.41,
1539
+ "learning_rate": 2.930591259640103e-05,
1540
+ "loss": 1.6486,
1541
+ "step": 550
1542
+ },
1543
+ {
1544
+ "epoch": 1.41,
1545
+ "eval_loss": 1.6600297689437866,
1546
+ "eval_runtime": 37.9635,
1547
+ "eval_samples_per_second": 2.213,
1548
+ "eval_steps_per_second": 0.29,
1549
+ "step": 550
1550
+ },
1551
+ {
1552
+ "epoch": 1.43,
1553
+ "learning_rate": 2.866323907455013e-05,
1554
+ "loss": 1.6861,
1555
+ "step": 555
1556
+ },
1557
+ {
1558
+ "epoch": 1.43,
1559
+ "eval_loss": 1.6602742671966553,
1560
+ "eval_runtime": 37.8722,
1561
+ "eval_samples_per_second": 2.218,
1562
+ "eval_steps_per_second": 0.29,
1563
+ "step": 555
1564
+ },
1565
+ {
1566
+ "epoch": 1.44,
1567
+ "learning_rate": 2.802056555269923e-05,
1568
+ "loss": 1.6624,
1569
+ "step": 560
1570
+ },
1571
+ {
1572
+ "epoch": 1.44,
1573
+ "eval_loss": 1.6599992513656616,
1574
+ "eval_runtime": 37.9306,
1575
+ "eval_samples_per_second": 2.215,
1576
+ "eval_steps_per_second": 0.29,
1577
+ "step": 560
1578
+ },
1579
+ {
1580
+ "epoch": 1.45,
1581
+ "learning_rate": 2.737789203084833e-05,
1582
+ "loss": 1.6323,
1583
+ "step": 565
1584
+ },
1585
+ {
1586
+ "epoch": 1.45,
1587
+ "eval_loss": 1.6600306034088135,
1588
+ "eval_runtime": 37.9548,
1589
+ "eval_samples_per_second": 2.213,
1590
+ "eval_steps_per_second": 0.29,
1591
+ "step": 565
1592
+ },
1593
+ {
1594
+ "epoch": 1.47,
1595
+ "learning_rate": 2.673521850899743e-05,
1596
+ "loss": 1.6707,
1597
+ "step": 570
1598
+ },
1599
+ {
1600
+ "epoch": 1.47,
1601
+ "eval_loss": 1.6601014137268066,
1602
+ "eval_runtime": 38.0349,
1603
+ "eval_samples_per_second": 2.208,
1604
+ "eval_steps_per_second": 0.289,
1605
+ "step": 570
1606
+ },
1607
+ {
1608
+ "epoch": 1.48,
1609
+ "learning_rate": 2.6092544987146534e-05,
1610
+ "loss": 1.6478,
1611
+ "step": 575
1612
+ },
1613
+ {
1614
+ "epoch": 1.48,
1615
+ "eval_loss": 1.6597967147827148,
1616
+ "eval_runtime": 37.9413,
1617
+ "eval_samples_per_second": 2.214,
1618
+ "eval_steps_per_second": 0.29,
1619
+ "step": 575
1620
+ },
1621
+ {
1622
+ "epoch": 1.49,
1623
+ "learning_rate": 2.5449871465295634e-05,
1624
+ "loss": 1.6715,
1625
+ "step": 580
1626
+ },
1627
+ {
1628
+ "epoch": 1.49,
1629
+ "eval_loss": 1.6598337888717651,
1630
+ "eval_runtime": 37.8386,
1631
+ "eval_samples_per_second": 2.22,
1632
+ "eval_steps_per_second": 0.291,
1633
+ "step": 580
1634
+ },
1635
+ {
1636
+ "epoch": 1.5,
1637
+ "learning_rate": 2.480719794344473e-05,
1638
+ "loss": 1.6626,
1639
+ "step": 585
1640
+ },
1641
+ {
1642
+ "epoch": 1.5,
1643
+ "eval_loss": 1.6600011587142944,
1644
+ "eval_runtime": 37.7901,
1645
+ "eval_samples_per_second": 2.223,
1646
+ "eval_steps_per_second": 0.291,
1647
+ "step": 585
1648
+ },
1649
+ {
1650
+ "epoch": 1.52,
1651
+ "learning_rate": 2.4164524421593832e-05,
1652
+ "loss": 1.651,
1653
+ "step": 590
1654
+ },
1655
+ {
1656
+ "epoch": 1.52,
1657
+ "eval_loss": 1.65969717502594,
1658
+ "eval_runtime": 37.7643,
1659
+ "eval_samples_per_second": 2.224,
1660
+ "eval_steps_per_second": 0.291,
1661
+ "step": 590
1662
+ },
1663
+ {
1664
+ "epoch": 1.53,
1665
+ "learning_rate": 2.3521850899742933e-05,
1666
+ "loss": 1.6346,
1667
+ "step": 595
1668
+ },
1669
+ {
1670
+ "epoch": 1.53,
1671
+ "eval_loss": 1.6598676443099976,
1672
+ "eval_runtime": 37.9026,
1673
+ "eval_samples_per_second": 2.216,
1674
+ "eval_steps_per_second": 0.29,
1675
+ "step": 595
1676
+ },
1677
+ {
1678
+ "epoch": 1.54,
1679
+ "learning_rate": 2.2879177377892033e-05,
1680
+ "loss": 1.6533,
1681
+ "step": 600
1682
+ },
1683
+ {
1684
+ "epoch": 1.54,
1685
+ "eval_loss": 1.659874677658081,
1686
+ "eval_runtime": 37.9871,
1687
+ "eval_samples_per_second": 2.211,
1688
+ "eval_steps_per_second": 0.29,
1689
+ "step": 600
1690
+ },
1691
+ {
1692
+ "epoch": 1.56,
1693
+ "learning_rate": 2.2236503856041134e-05,
1694
+ "loss": 1.6913,
1695
+ "step": 605
1696
+ },
1697
+ {
1698
+ "epoch": 1.56,
1699
+ "eval_loss": 1.659849762916565,
1700
+ "eval_runtime": 37.8712,
1701
+ "eval_samples_per_second": 2.218,
1702
+ "eval_steps_per_second": 0.29,
1703
+ "step": 605
1704
+ },
1705
+ {
1706
+ "epoch": 1.57,
1707
+ "learning_rate": 2.159383033419023e-05,
1708
+ "loss": 1.6709,
1709
+ "step": 610
1710
+ },
1711
+ {
1712
+ "epoch": 1.57,
1713
+ "eval_loss": 1.6598856449127197,
1714
+ "eval_runtime": 37.7043,
1715
+ "eval_samples_per_second": 2.228,
1716
+ "eval_steps_per_second": 0.292,
1717
+ "step": 610
1718
+ },
1719
+ {
1720
+ "epoch": 1.58,
1721
+ "learning_rate": 2.095115681233933e-05,
1722
+ "loss": 1.598,
1723
+ "step": 615
1724
+ },
1725
+ {
1726
+ "epoch": 1.58,
1727
+ "eval_loss": 1.6597983837127686,
1728
+ "eval_runtime": 37.722,
1729
+ "eval_samples_per_second": 2.227,
1730
+ "eval_steps_per_second": 0.292,
1731
+ "step": 615
1732
+ },
1733
+ {
1734
+ "epoch": 1.59,
1735
+ "learning_rate": 2.0308483290488432e-05,
1736
+ "loss": 1.6434,
1737
+ "step": 620
1738
+ },
1739
+ {
1740
+ "epoch": 1.59,
1741
+ "eval_loss": 1.6595433950424194,
1742
+ "eval_runtime": 37.6125,
1743
+ "eval_samples_per_second": 2.233,
1744
+ "eval_steps_per_second": 0.292,
1745
+ "step": 620
1746
+ },
1747
+ {
1748
+ "epoch": 1.61,
1749
+ "learning_rate": 1.9665809768637533e-05,
1750
+ "loss": 1.6086,
1751
+ "step": 625
1752
+ },
1753
+ {
1754
+ "epoch": 1.61,
1755
+ "eval_loss": 1.6594945192337036,
1756
+ "eval_runtime": 38.0046,
1757
+ "eval_samples_per_second": 2.21,
1758
+ "eval_steps_per_second": 0.289,
1759
+ "step": 625
1760
+ },
1761
+ {
1762
+ "epoch": 1.62,
1763
+ "learning_rate": 1.9023136246786633e-05,
1764
+ "loss": 1.5962,
1765
+ "step": 630
1766
+ },
1767
+ {
1768
+ "epoch": 1.62,
1769
+ "eval_loss": 1.6595605611801147,
1770
+ "eval_runtime": 37.8518,
1771
+ "eval_samples_per_second": 2.219,
1772
+ "eval_steps_per_second": 0.291,
1773
+ "step": 630
1774
+ },
1775
+ {
1776
+ "epoch": 1.63,
1777
+ "learning_rate": 1.8380462724935734e-05,
1778
+ "loss": 1.6695,
1779
+ "step": 635
1780
+ },
1781
+ {
1782
+ "epoch": 1.63,
1783
+ "eval_loss": 1.6592403650283813,
1784
+ "eval_runtime": 37.7285,
1785
+ "eval_samples_per_second": 2.226,
1786
+ "eval_steps_per_second": 0.292,
1787
+ "step": 635
1788
+ },
1789
+ {
1790
+ "epoch": 1.65,
1791
+ "learning_rate": 1.7737789203084834e-05,
1792
+ "loss": 1.6526,
1793
+ "step": 640
1794
+ },
1795
+ {
1796
+ "epoch": 1.65,
1797
+ "eval_loss": 1.6591880321502686,
1798
+ "eval_runtime": 37.758,
1799
+ "eval_samples_per_second": 2.225,
1800
+ "eval_steps_per_second": 0.291,
1801
+ "step": 640
1802
+ },
1803
+ {
1804
+ "epoch": 1.66,
1805
+ "learning_rate": 1.7095115681233935e-05,
1806
+ "loss": 1.672,
1807
+ "step": 645
1808
+ },
1809
+ {
1810
+ "epoch": 1.66,
1811
+ "eval_loss": 1.6588138341903687,
1812
+ "eval_runtime": 37.92,
1813
+ "eval_samples_per_second": 2.215,
1814
+ "eval_steps_per_second": 0.29,
1815
+ "step": 645
1816
+ },
1817
+ {
1818
+ "epoch": 1.67,
1819
+ "learning_rate": 1.6452442159383032e-05,
1820
+ "loss": 1.6347,
1821
+ "step": 650
1822
+ },
1823
+ {
1824
+ "epoch": 1.67,
1825
+ "eval_loss": 1.6589809656143188,
1826
+ "eval_runtime": 37.8021,
1827
+ "eval_samples_per_second": 2.222,
1828
+ "eval_steps_per_second": 0.291,
1829
+ "step": 650
1830
+ },
1831
+ {
1832
+ "epoch": 1.68,
1833
+ "learning_rate": 1.5809768637532136e-05,
1834
+ "loss": 1.6227,
1835
+ "step": 655
1836
+ },
1837
+ {
1838
+ "epoch": 1.68,
1839
+ "eval_loss": 1.6586792469024658,
1840
+ "eval_runtime": 37.9128,
1841
+ "eval_samples_per_second": 2.216,
1842
+ "eval_steps_per_second": 0.29,
1843
+ "step": 655
1844
+ },
1845
+ {
1846
+ "epoch": 1.7,
1847
+ "learning_rate": 1.5167095115681235e-05,
1848
+ "loss": 1.6308,
1849
+ "step": 660
1850
+ },
1851
+ {
1852
+ "epoch": 1.7,
1853
+ "eval_loss": 1.6585420370101929,
1854
+ "eval_runtime": 37.9881,
1855
+ "eval_samples_per_second": 2.211,
1856
+ "eval_steps_per_second": 0.29,
1857
+ "step": 660
1858
+ },
1859
+ {
1860
+ "epoch": 1.71,
1861
+ "learning_rate": 1.4524421593830334e-05,
1862
+ "loss": 1.6495,
1863
+ "step": 665
1864
+ },
1865
+ {
1866
+ "epoch": 1.71,
1867
+ "eval_loss": 1.6587530374526978,
1868
+ "eval_runtime": 37.9148,
1869
+ "eval_samples_per_second": 2.215,
1870
+ "eval_steps_per_second": 0.29,
1871
+ "step": 665
1872
+ },
1873
+ {
1874
+ "epoch": 1.72,
1875
+ "learning_rate": 1.3881748071979436e-05,
1876
+ "loss": 1.6712,
1877
+ "step": 670
1878
+ },
1879
+ {
1880
+ "epoch": 1.72,
1881
+ "eval_loss": 1.658648133277893,
1882
+ "eval_runtime": 37.9313,
1883
+ "eval_samples_per_second": 2.215,
1884
+ "eval_steps_per_second": 0.29,
1885
+ "step": 670
1886
+ },
1887
+ {
1888
+ "epoch": 1.74,
1889
+ "learning_rate": 1.3239074550128535e-05,
1890
+ "loss": 1.636,
1891
+ "step": 675
1892
+ },
1893
+ {
1894
+ "epoch": 1.74,
1895
+ "eval_loss": 1.6584407091140747,
1896
+ "eval_runtime": 37.7963,
1897
+ "eval_samples_per_second": 2.222,
1898
+ "eval_steps_per_second": 0.291,
1899
+ "step": 675
1900
+ },
1901
+ {
1902
+ "epoch": 1.75,
1903
+ "learning_rate": 1.2596401028277636e-05,
1904
+ "loss": 1.6453,
1905
+ "step": 680
1906
+ },
1907
+ {
1908
+ "epoch": 1.75,
1909
+ "eval_loss": 1.658594012260437,
1910
+ "eval_runtime": 37.7226,
1911
+ "eval_samples_per_second": 2.227,
1912
+ "eval_steps_per_second": 0.292,
1913
+ "step": 680
1914
+ },
1915
+ {
1916
+ "epoch": 1.76,
1917
+ "learning_rate": 1.1953727506426736e-05,
1918
+ "loss": 1.6509,
1919
+ "step": 685
1920
+ },
1921
+ {
1922
+ "epoch": 1.76,
1923
+ "eval_loss": 1.658467411994934,
1924
+ "eval_runtime": 37.8179,
1925
+ "eval_samples_per_second": 2.221,
1926
+ "eval_steps_per_second": 0.291,
1927
+ "step": 685
1928
+ },
1929
+ {
1930
+ "epoch": 1.77,
1931
+ "learning_rate": 1.1311053984575835e-05,
1932
+ "loss": 1.6489,
1933
+ "step": 690
1934
+ },
1935
+ {
1936
+ "epoch": 1.77,
1937
+ "eval_loss": 1.6585352420806885,
1938
+ "eval_runtime": 37.6768,
1939
+ "eval_samples_per_second": 2.229,
1940
+ "eval_steps_per_second": 0.292,
1941
+ "step": 690
1942
+ },
1943
+ {
1944
+ "epoch": 1.79,
1945
+ "learning_rate": 1.0668380462724936e-05,
1946
+ "loss": 1.6424,
1947
+ "step": 695
1948
+ },
1949
+ {
1950
+ "epoch": 1.79,
1951
+ "eval_loss": 1.6582835912704468,
1952
+ "eval_runtime": 38.0083,
1953
+ "eval_samples_per_second": 2.21,
1954
+ "eval_steps_per_second": 0.289,
1955
+ "step": 695
1956
+ },
1957
+ {
1958
+ "epoch": 1.8,
1959
+ "learning_rate": 1.0025706940874038e-05,
1960
+ "loss": 1.6131,
1961
+ "step": 700
1962
+ },
1963
+ {
1964
+ "epoch": 1.8,
1965
+ "eval_loss": 1.658105731010437,
1966
+ "eval_runtime": 37.8711,
1967
+ "eval_samples_per_second": 2.218,
1968
+ "eval_steps_per_second": 0.29,
1969
+ "step": 700
1970
+ },
1971
+ {
1972
+ "epoch": 1.81,
1973
+ "learning_rate": 9.383033419023137e-06,
1974
+ "loss": 1.6225,
1975
+ "step": 705
1976
+ },
1977
+ {
1978
+ "epoch": 1.81,
1979
+ "eval_loss": 1.658282995223999,
1980
+ "eval_runtime": 37.8125,
1981
+ "eval_samples_per_second": 2.221,
1982
+ "eval_steps_per_second": 0.291,
1983
+ "step": 705
1984
+ },
1985
+ {
1986
+ "epoch": 1.83,
1987
+ "learning_rate": 8.740359897172237e-06,
1988
+ "loss": 1.6928,
1989
+ "step": 710
1990
+ },
1991
+ {
1992
+ "epoch": 1.83,
1993
+ "eval_loss": 1.6581056118011475,
1994
+ "eval_runtime": 37.863,
1995
+ "eval_samples_per_second": 2.219,
1996
+ "eval_steps_per_second": 0.291,
1997
+ "step": 710
1998
+ },
1999
+ {
2000
+ "epoch": 1.84,
2001
+ "learning_rate": 8.097686375321336e-06,
2002
+ "loss": 1.6227,
2003
+ "step": 715
2004
+ },
2005
+ {
2006
+ "epoch": 1.84,
2007
+ "eval_loss": 1.6581188440322876,
2008
+ "eval_runtime": 37.8646,
2009
+ "eval_samples_per_second": 2.218,
2010
+ "eval_steps_per_second": 0.291,
2011
+ "step": 715
2012
+ },
2013
+ {
2014
+ "epoch": 1.85,
2015
+ "learning_rate": 7.4550128534704376e-06,
2016
+ "loss": 1.6593,
2017
+ "step": 720
2018
+ },
2019
+ {
2020
+ "epoch": 1.85,
2021
+ "eval_loss": 1.6579999923706055,
2022
+ "eval_runtime": 37.876,
2023
+ "eval_samples_per_second": 2.218,
2024
+ "eval_steps_per_second": 0.29,
2025
+ "step": 720
2026
+ },
2027
+ {
2028
+ "epoch": 1.86,
2029
+ "learning_rate": 6.812339331619537e-06,
2030
+ "loss": 1.6642,
2031
+ "step": 725
2032
+ },
2033
+ {
2034
+ "epoch": 1.86,
2035
+ "eval_loss": 1.65813410282135,
2036
+ "eval_runtime": 37.7093,
2037
+ "eval_samples_per_second": 2.228,
2038
+ "eval_steps_per_second": 0.292,
2039
+ "step": 725
2040
+ },
2041
+ {
2042
+ "epoch": 1.88,
2043
+ "learning_rate": 6.169665809768638e-06,
2044
+ "loss": 1.6414,
2045
+ "step": 730
2046
+ },
2047
+ {
2048
+ "epoch": 1.88,
2049
+ "eval_loss": 1.6581989526748657,
2050
+ "eval_runtime": 37.7818,
2051
+ "eval_samples_per_second": 2.223,
2052
+ "eval_steps_per_second": 0.291,
2053
+ "step": 730
2054
+ },
2055
+ {
2056
+ "epoch": 1.89,
2057
+ "learning_rate": 5.526992287917738e-06,
2058
+ "loss": 1.6519,
2059
+ "step": 735
2060
+ },
2061
+ {
2062
+ "epoch": 1.89,
2063
+ "eval_loss": 1.6577394008636475,
2064
+ "eval_runtime": 37.7264,
2065
+ "eval_samples_per_second": 2.227,
2066
+ "eval_steps_per_second": 0.292,
2067
+ "step": 735
2068
+ },
2069
+ {
2070
+ "epoch": 1.9,
2071
+ "learning_rate": 4.884318766066838e-06,
2072
+ "loss": 1.6485,
2073
+ "step": 740
2074
+ },
2075
+ {
2076
+ "epoch": 1.9,
2077
+ "eval_loss": 1.6580238342285156,
2078
+ "eval_runtime": 37.8729,
2079
+ "eval_samples_per_second": 2.218,
2080
+ "eval_steps_per_second": 0.29,
2081
+ "step": 740
2082
+ },
2083
+ {
2084
+ "epoch": 1.92,
2085
+ "learning_rate": 4.241645244215939e-06,
2086
+ "loss": 1.6261,
2087
+ "step": 745
2088
+ },
2089
+ {
2090
+ "epoch": 1.92,
2091
+ "eval_loss": 1.6580148935317993,
2092
+ "eval_runtime": 37.9546,
2093
+ "eval_samples_per_second": 2.213,
2094
+ "eval_steps_per_second": 0.29,
2095
+ "step": 745
2096
+ },
2097
+ {
2098
+ "epoch": 1.93,
2099
+ "learning_rate": 3.598971722365039e-06,
2100
+ "loss": 1.657,
2101
+ "step": 750
2102
+ },
2103
+ {
2104
+ "epoch": 1.93,
2105
+ "eval_loss": 1.657867670059204,
2106
+ "eval_runtime": 37.9411,
2107
+ "eval_samples_per_second": 2.214,
2108
+ "eval_steps_per_second": 0.29,
2109
+ "step": 750
2110
+ }
2111
+ ],
2112
+ "logging_steps": 5,
2113
+ "max_steps": 778,
2114
+ "num_train_epochs": 2,
2115
+ "save_steps": 10,
2116
+ "total_flos": 2.4364649130491904e+17,
2117
+ "trial_name": null,
2118
+ "trial_params": null
2119
+ }
checkpoint-750/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6211dd8a7344cda681d69cc75786fde8b7049338b76992777b83de12eab260a
3
+ size 4027
checkpoint-760/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: True
10
+ - load_in_4bit: False
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: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.5.0
checkpoint-760/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
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": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-760/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c49461f6486069a6430d2f9ef32aa1fa24fe7eaf8b85f00acb6ae38ef6eb04c4
3
+ size 16822989
checkpoint-760/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4033be186d9538808b2147fa85cbfe9b49759eba7d18d594cfba2658b548c17
3
+ size 33665221
checkpoint-760/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d07f69adc692dc0425df6c7ac98383b4abd1f6e8d926af37f8882a8c5bd2b28
3
+ size 14575
checkpoint-760/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e991e4ce437dc42eba99b838b76316581f55326eb30765cd80909c369a69ec1
3
+ size 627
checkpoint-760/trainer_state.json ADDED
@@ -0,0 +1,2147 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.6577980518341064,
3
+ "best_model_checkpoint": "/scratch/kwamea/llama-output/checkpoint-760",
4
+ "epoch": 1.9537275064267352,
5
+ "eval_steps": 5,
6
+ "global_step": 760,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 9.93573264781491e-05,
14
+ "loss": 1.9824,
15
+ "step": 5
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "eval_loss": 1.954480528831482,
20
+ "eval_runtime": 38.0417,
21
+ "eval_samples_per_second": 2.208,
22
+ "eval_steps_per_second": 0.289,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.03,
27
+ "learning_rate": 9.87146529562982e-05,
28
+ "loss": 1.9245,
29
+ "step": 10
30
+ },
31
+ {
32
+ "epoch": 0.03,
33
+ "eval_loss": 1.8942010402679443,
34
+ "eval_runtime": 37.8934,
35
+ "eval_samples_per_second": 2.217,
36
+ "eval_steps_per_second": 0.29,
37
+ "step": 10
38
+ },
39
+ {
40
+ "epoch": 0.04,
41
+ "learning_rate": 9.80719794344473e-05,
42
+ "loss": 1.8616,
43
+ "step": 15
44
+ },
45
+ {
46
+ "epoch": 0.04,
47
+ "eval_loss": 1.8424055576324463,
48
+ "eval_runtime": 37.8727,
49
+ "eval_samples_per_second": 2.218,
50
+ "eval_steps_per_second": 0.29,
51
+ "step": 15
52
+ },
53
+ {
54
+ "epoch": 0.05,
55
+ "learning_rate": 9.742930591259641e-05,
56
+ "loss": 1.8267,
57
+ "step": 20
58
+ },
59
+ {
60
+ "epoch": 0.05,
61
+ "eval_loss": 1.8279685974121094,
62
+ "eval_runtime": 38.166,
63
+ "eval_samples_per_second": 2.201,
64
+ "eval_steps_per_second": 0.288,
65
+ "step": 20
66
+ },
67
+ {
68
+ "epoch": 0.06,
69
+ "learning_rate": 9.67866323907455e-05,
70
+ "loss": 1.835,
71
+ "step": 25
72
+ },
73
+ {
74
+ "epoch": 0.06,
75
+ "eval_loss": 1.8134318590164185,
76
+ "eval_runtime": 38.0411,
77
+ "eval_samples_per_second": 2.208,
78
+ "eval_steps_per_second": 0.289,
79
+ "step": 25
80
+ },
81
+ {
82
+ "epoch": 0.08,
83
+ "learning_rate": 9.61439588688946e-05,
84
+ "loss": 1.8236,
85
+ "step": 30
86
+ },
87
+ {
88
+ "epoch": 0.08,
89
+ "eval_loss": 1.8010995388031006,
90
+ "eval_runtime": 37.8753,
91
+ "eval_samples_per_second": 2.218,
92
+ "eval_steps_per_second": 0.29,
93
+ "step": 30
94
+ },
95
+ {
96
+ "epoch": 0.09,
97
+ "learning_rate": 9.550128534704372e-05,
98
+ "loss": 1.7186,
99
+ "step": 35
100
+ },
101
+ {
102
+ "epoch": 0.09,
103
+ "eval_loss": 1.7901933193206787,
104
+ "eval_runtime": 38.0374,
105
+ "eval_samples_per_second": 2.208,
106
+ "eval_steps_per_second": 0.289,
107
+ "step": 35
108
+ },
109
+ {
110
+ "epoch": 0.1,
111
+ "learning_rate": 9.485861182519281e-05,
112
+ "loss": 1.7959,
113
+ "step": 40
114
+ },
115
+ {
116
+ "epoch": 0.1,
117
+ "eval_loss": 1.7807564735412598,
118
+ "eval_runtime": 37.8784,
119
+ "eval_samples_per_second": 2.218,
120
+ "eval_steps_per_second": 0.29,
121
+ "step": 40
122
+ },
123
+ {
124
+ "epoch": 0.12,
125
+ "learning_rate": 9.421593830334192e-05,
126
+ "loss": 1.715,
127
+ "step": 45
128
+ },
129
+ {
130
+ "epoch": 0.12,
131
+ "eval_loss": 1.7709146738052368,
132
+ "eval_runtime": 37.8705,
133
+ "eval_samples_per_second": 2.218,
134
+ "eval_steps_per_second": 0.29,
135
+ "step": 45
136
+ },
137
+ {
138
+ "epoch": 0.13,
139
+ "learning_rate": 9.357326478149101e-05,
140
+ "loss": 1.7581,
141
+ "step": 50
142
+ },
143
+ {
144
+ "epoch": 0.13,
145
+ "eval_loss": 1.761339783668518,
146
+ "eval_runtime": 37.661,
147
+ "eval_samples_per_second": 2.23,
148
+ "eval_steps_per_second": 0.292,
149
+ "step": 50
150
+ },
151
+ {
152
+ "epoch": 0.14,
153
+ "learning_rate": 9.29305912596401e-05,
154
+ "loss": 1.7305,
155
+ "step": 55
156
+ },
157
+ {
158
+ "epoch": 0.14,
159
+ "eval_loss": 1.7389689683914185,
160
+ "eval_runtime": 38.0055,
161
+ "eval_samples_per_second": 2.21,
162
+ "eval_steps_per_second": 0.289,
163
+ "step": 55
164
+ },
165
+ {
166
+ "epoch": 0.15,
167
+ "learning_rate": 9.228791773778921e-05,
168
+ "loss": 1.7086,
169
+ "step": 60
170
+ },
171
+ {
172
+ "epoch": 0.15,
173
+ "eval_loss": 1.713218092918396,
174
+ "eval_runtime": 37.737,
175
+ "eval_samples_per_second": 2.226,
176
+ "eval_steps_per_second": 0.291,
177
+ "step": 60
178
+ },
179
+ {
180
+ "epoch": 0.17,
181
+ "learning_rate": 9.16452442159383e-05,
182
+ "loss": 1.7057,
183
+ "step": 65
184
+ },
185
+ {
186
+ "epoch": 0.17,
187
+ "eval_loss": 1.7108122110366821,
188
+ "eval_runtime": 38.0037,
189
+ "eval_samples_per_second": 2.21,
190
+ "eval_steps_per_second": 0.289,
191
+ "step": 65
192
+ },
193
+ {
194
+ "epoch": 0.18,
195
+ "learning_rate": 9.100257069408741e-05,
196
+ "loss": 1.7243,
197
+ "step": 70
198
+ },
199
+ {
200
+ "epoch": 0.18,
201
+ "eval_loss": 1.7041622400283813,
202
+ "eval_runtime": 38.1402,
203
+ "eval_samples_per_second": 2.202,
204
+ "eval_steps_per_second": 0.288,
205
+ "step": 70
206
+ },
207
+ {
208
+ "epoch": 0.19,
209
+ "learning_rate": 9.03598971722365e-05,
210
+ "loss": 1.7038,
211
+ "step": 75
212
+ },
213
+ {
214
+ "epoch": 0.19,
215
+ "eval_loss": 1.7010408639907837,
216
+ "eval_runtime": 37.7193,
217
+ "eval_samples_per_second": 2.227,
218
+ "eval_steps_per_second": 0.292,
219
+ "step": 75
220
+ },
221
+ {
222
+ "epoch": 0.21,
223
+ "learning_rate": 8.97172236503856e-05,
224
+ "loss": 1.6939,
225
+ "step": 80
226
+ },
227
+ {
228
+ "epoch": 0.21,
229
+ "eval_loss": 1.6987501382827759,
230
+ "eval_runtime": 38.0624,
231
+ "eval_samples_per_second": 2.207,
232
+ "eval_steps_per_second": 0.289,
233
+ "step": 80
234
+ },
235
+ {
236
+ "epoch": 0.22,
237
+ "learning_rate": 8.907455012853471e-05,
238
+ "loss": 1.645,
239
+ "step": 85
240
+ },
241
+ {
242
+ "epoch": 0.22,
243
+ "eval_loss": 1.6968218088150024,
244
+ "eval_runtime": 37.9205,
245
+ "eval_samples_per_second": 2.215,
246
+ "eval_steps_per_second": 0.29,
247
+ "step": 85
248
+ },
249
+ {
250
+ "epoch": 0.23,
251
+ "learning_rate": 8.84318766066838e-05,
252
+ "loss": 1.7016,
253
+ "step": 90
254
+ },
255
+ {
256
+ "epoch": 0.23,
257
+ "eval_loss": 1.6952500343322754,
258
+ "eval_runtime": 37.9617,
259
+ "eval_samples_per_second": 2.213,
260
+ "eval_steps_per_second": 0.29,
261
+ "step": 90
262
+ },
263
+ {
264
+ "epoch": 0.24,
265
+ "learning_rate": 8.778920308483291e-05,
266
+ "loss": 1.6869,
267
+ "step": 95
268
+ },
269
+ {
270
+ "epoch": 0.24,
271
+ "eval_loss": 1.6932854652404785,
272
+ "eval_runtime": 37.8781,
273
+ "eval_samples_per_second": 2.218,
274
+ "eval_steps_per_second": 0.29,
275
+ "step": 95
276
+ },
277
+ {
278
+ "epoch": 0.26,
279
+ "learning_rate": 8.7146529562982e-05,
280
+ "loss": 1.6728,
281
+ "step": 100
282
+ },
283
+ {
284
+ "epoch": 0.26,
285
+ "eval_loss": 1.6920738220214844,
286
+ "eval_runtime": 38.1996,
287
+ "eval_samples_per_second": 2.199,
288
+ "eval_steps_per_second": 0.288,
289
+ "step": 100
290
+ },
291
+ {
292
+ "epoch": 0.27,
293
+ "learning_rate": 8.650385604113111e-05,
294
+ "loss": 1.6674,
295
+ "step": 105
296
+ },
297
+ {
298
+ "epoch": 0.27,
299
+ "eval_loss": 1.6904469728469849,
300
+ "eval_runtime": 38.1252,
301
+ "eval_samples_per_second": 2.203,
302
+ "eval_steps_per_second": 0.289,
303
+ "step": 105
304
+ },
305
+ {
306
+ "epoch": 0.28,
307
+ "learning_rate": 8.586118251928022e-05,
308
+ "loss": 1.7013,
309
+ "step": 110
310
+ },
311
+ {
312
+ "epoch": 0.28,
313
+ "eval_loss": 1.6895124912261963,
314
+ "eval_runtime": 38.0968,
315
+ "eval_samples_per_second": 2.205,
316
+ "eval_steps_per_second": 0.289,
317
+ "step": 110
318
+ },
319
+ {
320
+ "epoch": 0.3,
321
+ "learning_rate": 8.521850899742931e-05,
322
+ "loss": 1.7148,
323
+ "step": 115
324
+ },
325
+ {
326
+ "epoch": 0.3,
327
+ "eval_loss": 1.6886374950408936,
328
+ "eval_runtime": 37.8799,
329
+ "eval_samples_per_second": 2.218,
330
+ "eval_steps_per_second": 0.29,
331
+ "step": 115
332
+ },
333
+ {
334
+ "epoch": 0.31,
335
+ "learning_rate": 8.457583547557842e-05,
336
+ "loss": 1.7166,
337
+ "step": 120
338
+ },
339
+ {
340
+ "epoch": 0.31,
341
+ "eval_loss": 1.6868164539337158,
342
+ "eval_runtime": 37.8861,
343
+ "eval_samples_per_second": 2.217,
344
+ "eval_steps_per_second": 0.29,
345
+ "step": 120
346
+ },
347
+ {
348
+ "epoch": 0.32,
349
+ "learning_rate": 8.393316195372751e-05,
350
+ "loss": 1.7012,
351
+ "step": 125
352
+ },
353
+ {
354
+ "epoch": 0.32,
355
+ "eval_loss": 1.6858367919921875,
356
+ "eval_runtime": 38.1136,
357
+ "eval_samples_per_second": 2.204,
358
+ "eval_steps_per_second": 0.289,
359
+ "step": 125
360
+ },
361
+ {
362
+ "epoch": 0.33,
363
+ "learning_rate": 8.32904884318766e-05,
364
+ "loss": 1.6827,
365
+ "step": 130
366
+ },
367
+ {
368
+ "epoch": 0.33,
369
+ "eval_loss": 1.6849361658096313,
370
+ "eval_runtime": 37.9883,
371
+ "eval_samples_per_second": 2.211,
372
+ "eval_steps_per_second": 0.29,
373
+ "step": 130
374
+ },
375
+ {
376
+ "epoch": 0.35,
377
+ "learning_rate": 8.264781491002571e-05,
378
+ "loss": 1.7157,
379
+ "step": 135
380
+ },
381
+ {
382
+ "epoch": 0.35,
383
+ "eval_loss": 1.684756875038147,
384
+ "eval_runtime": 37.8726,
385
+ "eval_samples_per_second": 2.218,
386
+ "eval_steps_per_second": 0.29,
387
+ "step": 135
388
+ },
389
+ {
390
+ "epoch": 0.36,
391
+ "learning_rate": 8.200514138817481e-05,
392
+ "loss": 1.6668,
393
+ "step": 140
394
+ },
395
+ {
396
+ "epoch": 0.36,
397
+ "eval_loss": 1.6836236715316772,
398
+ "eval_runtime": 37.9586,
399
+ "eval_samples_per_second": 2.213,
400
+ "eval_steps_per_second": 0.29,
401
+ "step": 140
402
+ },
403
+ {
404
+ "epoch": 0.37,
405
+ "learning_rate": 8.136246786632391e-05,
406
+ "loss": 1.6588,
407
+ "step": 145
408
+ },
409
+ {
410
+ "epoch": 0.37,
411
+ "eval_loss": 1.6828166246414185,
412
+ "eval_runtime": 38.0596,
413
+ "eval_samples_per_second": 2.207,
414
+ "eval_steps_per_second": 0.289,
415
+ "step": 145
416
+ },
417
+ {
418
+ "epoch": 0.39,
419
+ "learning_rate": 8.071979434447301e-05,
420
+ "loss": 1.7005,
421
+ "step": 150
422
+ },
423
+ {
424
+ "epoch": 0.39,
425
+ "eval_loss": 1.681463360786438,
426
+ "eval_runtime": 38.0878,
427
+ "eval_samples_per_second": 2.205,
428
+ "eval_steps_per_second": 0.289,
429
+ "step": 150
430
+ },
431
+ {
432
+ "epoch": 0.4,
433
+ "learning_rate": 8.007712082262212e-05,
434
+ "loss": 1.6893,
435
+ "step": 155
436
+ },
437
+ {
438
+ "epoch": 0.4,
439
+ "eval_loss": 1.6811552047729492,
440
+ "eval_runtime": 37.9265,
441
+ "eval_samples_per_second": 2.215,
442
+ "eval_steps_per_second": 0.29,
443
+ "step": 155
444
+ },
445
+ {
446
+ "epoch": 0.41,
447
+ "learning_rate": 7.943444730077121e-05,
448
+ "loss": 1.6682,
449
+ "step": 160
450
+ },
451
+ {
452
+ "epoch": 0.41,
453
+ "eval_loss": 1.6801096200942993,
454
+ "eval_runtime": 38.0375,
455
+ "eval_samples_per_second": 2.208,
456
+ "eval_steps_per_second": 0.289,
457
+ "step": 160
458
+ },
459
+ {
460
+ "epoch": 0.42,
461
+ "learning_rate": 7.87917737789203e-05,
462
+ "loss": 1.6481,
463
+ "step": 165
464
+ },
465
+ {
466
+ "epoch": 0.42,
467
+ "eval_loss": 1.6800144910812378,
468
+ "eval_runtime": 37.8916,
469
+ "eval_samples_per_second": 2.217,
470
+ "eval_steps_per_second": 0.29,
471
+ "step": 165
472
+ },
473
+ {
474
+ "epoch": 0.44,
475
+ "learning_rate": 7.814910025706941e-05,
476
+ "loss": 1.7042,
477
+ "step": 170
478
+ },
479
+ {
480
+ "epoch": 0.44,
481
+ "eval_loss": 1.6784976720809937,
482
+ "eval_runtime": 37.8514,
483
+ "eval_samples_per_second": 2.219,
484
+ "eval_steps_per_second": 0.291,
485
+ "step": 170
486
+ },
487
+ {
488
+ "epoch": 0.45,
489
+ "learning_rate": 7.750642673521852e-05,
490
+ "loss": 1.6555,
491
+ "step": 175
492
+ },
493
+ {
494
+ "epoch": 0.45,
495
+ "eval_loss": 1.6780468225479126,
496
+ "eval_runtime": 37.8893,
497
+ "eval_samples_per_second": 2.217,
498
+ "eval_steps_per_second": 0.29,
499
+ "step": 175
500
+ },
501
+ {
502
+ "epoch": 0.46,
503
+ "learning_rate": 7.686375321336761e-05,
504
+ "loss": 1.6717,
505
+ "step": 180
506
+ },
507
+ {
508
+ "epoch": 0.46,
509
+ "eval_loss": 1.6775085926055908,
510
+ "eval_runtime": 37.9289,
511
+ "eval_samples_per_second": 2.215,
512
+ "eval_steps_per_second": 0.29,
513
+ "step": 180
514
+ },
515
+ {
516
+ "epoch": 0.48,
517
+ "learning_rate": 7.622107969151672e-05,
518
+ "loss": 1.6716,
519
+ "step": 185
520
+ },
521
+ {
522
+ "epoch": 0.48,
523
+ "eval_loss": 1.6770914793014526,
524
+ "eval_runtime": 38.0111,
525
+ "eval_samples_per_second": 2.21,
526
+ "eval_steps_per_second": 0.289,
527
+ "step": 185
528
+ },
529
+ {
530
+ "epoch": 0.49,
531
+ "learning_rate": 7.557840616966581e-05,
532
+ "loss": 1.6618,
533
+ "step": 190
534
+ },
535
+ {
536
+ "epoch": 0.49,
537
+ "eval_loss": 1.676563024520874,
538
+ "eval_runtime": 38.0703,
539
+ "eval_samples_per_second": 2.206,
540
+ "eval_steps_per_second": 0.289,
541
+ "step": 190
542
+ },
543
+ {
544
+ "epoch": 0.5,
545
+ "learning_rate": 7.493573264781492e-05,
546
+ "loss": 1.6411,
547
+ "step": 195
548
+ },
549
+ {
550
+ "epoch": 0.5,
551
+ "eval_loss": 1.6757832765579224,
552
+ "eval_runtime": 37.827,
553
+ "eval_samples_per_second": 2.221,
554
+ "eval_steps_per_second": 0.291,
555
+ "step": 195
556
+ },
557
+ {
558
+ "epoch": 0.51,
559
+ "learning_rate": 7.429305912596401e-05,
560
+ "loss": 1.6637,
561
+ "step": 200
562
+ },
563
+ {
564
+ "epoch": 0.51,
565
+ "eval_loss": 1.6751985549926758,
566
+ "eval_runtime": 37.836,
567
+ "eval_samples_per_second": 2.22,
568
+ "eval_steps_per_second": 0.291,
569
+ "step": 200
570
+ },
571
+ {
572
+ "epoch": 0.53,
573
+ "learning_rate": 7.365038560411311e-05,
574
+ "loss": 1.6142,
575
+ "step": 205
576
+ },
577
+ {
578
+ "epoch": 0.53,
579
+ "eval_loss": 1.675147294998169,
580
+ "eval_runtime": 37.9175,
581
+ "eval_samples_per_second": 2.215,
582
+ "eval_steps_per_second": 0.29,
583
+ "step": 205
584
+ },
585
+ {
586
+ "epoch": 0.54,
587
+ "learning_rate": 7.300771208226222e-05,
588
+ "loss": 1.6741,
589
+ "step": 210
590
+ },
591
+ {
592
+ "epoch": 0.54,
593
+ "eval_loss": 1.674437165260315,
594
+ "eval_runtime": 38.087,
595
+ "eval_samples_per_second": 2.205,
596
+ "eval_steps_per_second": 0.289,
597
+ "step": 210
598
+ },
599
+ {
600
+ "epoch": 0.55,
601
+ "learning_rate": 7.236503856041131e-05,
602
+ "loss": 1.6408,
603
+ "step": 215
604
+ },
605
+ {
606
+ "epoch": 0.55,
607
+ "eval_loss": 1.6737719774246216,
608
+ "eval_runtime": 38.0075,
609
+ "eval_samples_per_second": 2.21,
610
+ "eval_steps_per_second": 0.289,
611
+ "step": 215
612
+ },
613
+ {
614
+ "epoch": 0.57,
615
+ "learning_rate": 7.172236503856042e-05,
616
+ "loss": 1.6733,
617
+ "step": 220
618
+ },
619
+ {
620
+ "epoch": 0.57,
621
+ "eval_loss": 1.6732759475708008,
622
+ "eval_runtime": 38.1125,
623
+ "eval_samples_per_second": 2.204,
624
+ "eval_steps_per_second": 0.289,
625
+ "step": 220
626
+ },
627
+ {
628
+ "epoch": 0.58,
629
+ "learning_rate": 7.107969151670951e-05,
630
+ "loss": 1.679,
631
+ "step": 225
632
+ },
633
+ {
634
+ "epoch": 0.58,
635
+ "eval_loss": 1.6726195812225342,
636
+ "eval_runtime": 37.9769,
637
+ "eval_samples_per_second": 2.212,
638
+ "eval_steps_per_second": 0.29,
639
+ "step": 225
640
+ },
641
+ {
642
+ "epoch": 0.59,
643
+ "learning_rate": 7.043701799485862e-05,
644
+ "loss": 1.7202,
645
+ "step": 230
646
+ },
647
+ {
648
+ "epoch": 0.59,
649
+ "eval_loss": 1.671908974647522,
650
+ "eval_runtime": 37.8215,
651
+ "eval_samples_per_second": 2.221,
652
+ "eval_steps_per_second": 0.291,
653
+ "step": 230
654
+ },
655
+ {
656
+ "epoch": 0.6,
657
+ "learning_rate": 6.979434447300771e-05,
658
+ "loss": 1.6805,
659
+ "step": 235
660
+ },
661
+ {
662
+ "epoch": 0.6,
663
+ "eval_loss": 1.6715577840805054,
664
+ "eval_runtime": 37.9343,
665
+ "eval_samples_per_second": 2.214,
666
+ "eval_steps_per_second": 0.29,
667
+ "step": 235
668
+ },
669
+ {
670
+ "epoch": 0.62,
671
+ "learning_rate": 6.91516709511568e-05,
672
+ "loss": 1.6331,
673
+ "step": 240
674
+ },
675
+ {
676
+ "epoch": 0.62,
677
+ "eval_loss": 1.6715376377105713,
678
+ "eval_runtime": 37.9904,
679
+ "eval_samples_per_second": 2.211,
680
+ "eval_steps_per_second": 0.29,
681
+ "step": 240
682
+ },
683
+ {
684
+ "epoch": 0.63,
685
+ "learning_rate": 6.850899742930593e-05,
686
+ "loss": 1.6761,
687
+ "step": 245
688
+ },
689
+ {
690
+ "epoch": 0.63,
691
+ "eval_loss": 1.671446681022644,
692
+ "eval_runtime": 38.0357,
693
+ "eval_samples_per_second": 2.208,
694
+ "eval_steps_per_second": 0.289,
695
+ "step": 245
696
+ },
697
+ {
698
+ "epoch": 0.64,
699
+ "learning_rate": 6.786632390745502e-05,
700
+ "loss": 1.6994,
701
+ "step": 250
702
+ },
703
+ {
704
+ "epoch": 0.64,
705
+ "eval_loss": 1.6716604232788086,
706
+ "eval_runtime": 37.9609,
707
+ "eval_samples_per_second": 2.213,
708
+ "eval_steps_per_second": 0.29,
709
+ "step": 250
710
+ },
711
+ {
712
+ "epoch": 0.66,
713
+ "learning_rate": 6.722365038560411e-05,
714
+ "loss": 1.6305,
715
+ "step": 255
716
+ },
717
+ {
718
+ "epoch": 0.66,
719
+ "eval_loss": 1.6711723804473877,
720
+ "eval_runtime": 37.7893,
721
+ "eval_samples_per_second": 2.223,
722
+ "eval_steps_per_second": 0.291,
723
+ "step": 255
724
+ },
725
+ {
726
+ "epoch": 0.67,
727
+ "learning_rate": 6.658097686375322e-05,
728
+ "loss": 1.6612,
729
+ "step": 260
730
+ },
731
+ {
732
+ "epoch": 0.67,
733
+ "eval_loss": 1.670398473739624,
734
+ "eval_runtime": 37.7667,
735
+ "eval_samples_per_second": 2.224,
736
+ "eval_steps_per_second": 0.291,
737
+ "step": 260
738
+ },
739
+ {
740
+ "epoch": 0.68,
741
+ "learning_rate": 6.593830334190231e-05,
742
+ "loss": 1.6576,
743
+ "step": 265
744
+ },
745
+ {
746
+ "epoch": 0.68,
747
+ "eval_loss": 1.6706459522247314,
748
+ "eval_runtime": 37.8285,
749
+ "eval_samples_per_second": 2.221,
750
+ "eval_steps_per_second": 0.291,
751
+ "step": 265
752
+ },
753
+ {
754
+ "epoch": 0.69,
755
+ "learning_rate": 6.529562982005142e-05,
756
+ "loss": 1.6837,
757
+ "step": 270
758
+ },
759
+ {
760
+ "epoch": 0.69,
761
+ "eval_loss": 1.6699285507202148,
762
+ "eval_runtime": 37.7619,
763
+ "eval_samples_per_second": 2.224,
764
+ "eval_steps_per_second": 0.291,
765
+ "step": 270
766
+ },
767
+ {
768
+ "epoch": 0.71,
769
+ "learning_rate": 6.465295629820052e-05,
770
+ "loss": 1.6493,
771
+ "step": 275
772
+ },
773
+ {
774
+ "epoch": 0.71,
775
+ "eval_loss": 1.6693135499954224,
776
+ "eval_runtime": 37.7441,
777
+ "eval_samples_per_second": 2.226,
778
+ "eval_steps_per_second": 0.291,
779
+ "step": 275
780
+ },
781
+ {
782
+ "epoch": 0.72,
783
+ "learning_rate": 6.401028277634962e-05,
784
+ "loss": 1.6427,
785
+ "step": 280
786
+ },
787
+ {
788
+ "epoch": 0.72,
789
+ "eval_loss": 1.669467568397522,
790
+ "eval_runtime": 37.8707,
791
+ "eval_samples_per_second": 2.218,
792
+ "eval_steps_per_second": 0.29,
793
+ "step": 280
794
+ },
795
+ {
796
+ "epoch": 0.73,
797
+ "learning_rate": 6.336760925449872e-05,
798
+ "loss": 1.6236,
799
+ "step": 285
800
+ },
801
+ {
802
+ "epoch": 0.73,
803
+ "eval_loss": 1.6690431833267212,
804
+ "eval_runtime": 37.8946,
805
+ "eval_samples_per_second": 2.217,
806
+ "eval_steps_per_second": 0.29,
807
+ "step": 285
808
+ },
809
+ {
810
+ "epoch": 0.75,
811
+ "learning_rate": 6.272493573264781e-05,
812
+ "loss": 1.6538,
813
+ "step": 290
814
+ },
815
+ {
816
+ "epoch": 0.75,
817
+ "eval_loss": 1.6688117980957031,
818
+ "eval_runtime": 37.9378,
819
+ "eval_samples_per_second": 2.214,
820
+ "eval_steps_per_second": 0.29,
821
+ "step": 290
822
+ },
823
+ {
824
+ "epoch": 0.76,
825
+ "learning_rate": 6.208226221079692e-05,
826
+ "loss": 1.6558,
827
+ "step": 295
828
+ },
829
+ {
830
+ "epoch": 0.76,
831
+ "eval_loss": 1.668560266494751,
832
+ "eval_runtime": 38.0448,
833
+ "eval_samples_per_second": 2.208,
834
+ "eval_steps_per_second": 0.289,
835
+ "step": 295
836
+ },
837
+ {
838
+ "epoch": 0.77,
839
+ "learning_rate": 6.143958868894601e-05,
840
+ "loss": 1.6401,
841
+ "step": 300
842
+ },
843
+ {
844
+ "epoch": 0.77,
845
+ "eval_loss": 1.6680612564086914,
846
+ "eval_runtime": 37.8571,
847
+ "eval_samples_per_second": 2.219,
848
+ "eval_steps_per_second": 0.291,
849
+ "step": 300
850
+ },
851
+ {
852
+ "epoch": 0.78,
853
+ "learning_rate": 6.079691516709511e-05,
854
+ "loss": 1.6468,
855
+ "step": 305
856
+ },
857
+ {
858
+ "epoch": 0.78,
859
+ "eval_loss": 1.6675914525985718,
860
+ "eval_runtime": 37.8491,
861
+ "eval_samples_per_second": 2.219,
862
+ "eval_steps_per_second": 0.291,
863
+ "step": 305
864
+ },
865
+ {
866
+ "epoch": 0.8,
867
+ "learning_rate": 6.015424164524421e-05,
868
+ "loss": 1.6579,
869
+ "step": 310
870
+ },
871
+ {
872
+ "epoch": 0.8,
873
+ "eval_loss": 1.6671632528305054,
874
+ "eval_runtime": 37.7658,
875
+ "eval_samples_per_second": 2.224,
876
+ "eval_steps_per_second": 0.291,
877
+ "step": 310
878
+ },
879
+ {
880
+ "epoch": 0.81,
881
+ "learning_rate": 5.951156812339333e-05,
882
+ "loss": 1.6339,
883
+ "step": 315
884
+ },
885
+ {
886
+ "epoch": 0.81,
887
+ "eval_loss": 1.666803240776062,
888
+ "eval_runtime": 37.7896,
889
+ "eval_samples_per_second": 2.223,
890
+ "eval_steps_per_second": 0.291,
891
+ "step": 315
892
+ },
893
+ {
894
+ "epoch": 0.82,
895
+ "learning_rate": 5.886889460154242e-05,
896
+ "loss": 1.6636,
897
+ "step": 320
898
+ },
899
+ {
900
+ "epoch": 0.82,
901
+ "eval_loss": 1.6664105653762817,
902
+ "eval_runtime": 38.0586,
903
+ "eval_samples_per_second": 2.207,
904
+ "eval_steps_per_second": 0.289,
905
+ "step": 320
906
+ },
907
+ {
908
+ "epoch": 0.84,
909
+ "learning_rate": 5.822622107969152e-05,
910
+ "loss": 1.641,
911
+ "step": 325
912
+ },
913
+ {
914
+ "epoch": 0.84,
915
+ "eval_loss": 1.6665875911712646,
916
+ "eval_runtime": 37.9046,
917
+ "eval_samples_per_second": 2.216,
918
+ "eval_steps_per_second": 0.29,
919
+ "step": 325
920
+ },
921
+ {
922
+ "epoch": 0.85,
923
+ "learning_rate": 5.758354755784062e-05,
924
+ "loss": 1.6616,
925
+ "step": 330
926
+ },
927
+ {
928
+ "epoch": 0.85,
929
+ "eval_loss": 1.6664165258407593,
930
+ "eval_runtime": 37.8575,
931
+ "eval_samples_per_second": 2.219,
932
+ "eval_steps_per_second": 0.291,
933
+ "step": 330
934
+ },
935
+ {
936
+ "epoch": 0.86,
937
+ "learning_rate": 5.694087403598972e-05,
938
+ "loss": 1.6597,
939
+ "step": 335
940
+ },
941
+ {
942
+ "epoch": 0.86,
943
+ "eval_loss": 1.666045904159546,
944
+ "eval_runtime": 37.9711,
945
+ "eval_samples_per_second": 2.212,
946
+ "eval_steps_per_second": 0.29,
947
+ "step": 335
948
+ },
949
+ {
950
+ "epoch": 0.87,
951
+ "learning_rate": 5.6298200514138824e-05,
952
+ "loss": 1.67,
953
+ "step": 340
954
+ },
955
+ {
956
+ "epoch": 0.87,
957
+ "eval_loss": 1.666337490081787,
958
+ "eval_runtime": 37.8821,
959
+ "eval_samples_per_second": 2.217,
960
+ "eval_steps_per_second": 0.29,
961
+ "step": 340
962
+ },
963
+ {
964
+ "epoch": 0.89,
965
+ "learning_rate": 5.5655526992287924e-05,
966
+ "loss": 1.6344,
967
+ "step": 345
968
+ },
969
+ {
970
+ "epoch": 0.89,
971
+ "eval_loss": 1.6659029722213745,
972
+ "eval_runtime": 37.9188,
973
+ "eval_samples_per_second": 2.215,
974
+ "eval_steps_per_second": 0.29,
975
+ "step": 345
976
+ },
977
+ {
978
+ "epoch": 0.9,
979
+ "learning_rate": 5.501285347043702e-05,
980
+ "loss": 1.6623,
981
+ "step": 350
982
+ },
983
+ {
984
+ "epoch": 0.9,
985
+ "eval_loss": 1.6656525135040283,
986
+ "eval_runtime": 37.9951,
987
+ "eval_samples_per_second": 2.211,
988
+ "eval_steps_per_second": 0.29,
989
+ "step": 350
990
+ },
991
+ {
992
+ "epoch": 0.91,
993
+ "learning_rate": 5.437017994858612e-05,
994
+ "loss": 1.6623,
995
+ "step": 355
996
+ },
997
+ {
998
+ "epoch": 0.91,
999
+ "eval_loss": 1.6654421091079712,
1000
+ "eval_runtime": 37.8803,
1001
+ "eval_samples_per_second": 2.218,
1002
+ "eval_steps_per_second": 0.29,
1003
+ "step": 355
1004
+ },
1005
+ {
1006
+ "epoch": 0.93,
1007
+ "learning_rate": 5.372750642673522e-05,
1008
+ "loss": 1.6741,
1009
+ "step": 360
1010
+ },
1011
+ {
1012
+ "epoch": 0.93,
1013
+ "eval_loss": 1.665541648864746,
1014
+ "eval_runtime": 37.7151,
1015
+ "eval_samples_per_second": 2.227,
1016
+ "eval_steps_per_second": 0.292,
1017
+ "step": 360
1018
+ },
1019
+ {
1020
+ "epoch": 0.94,
1021
+ "learning_rate": 5.308483290488432e-05,
1022
+ "loss": 1.6973,
1023
+ "step": 365
1024
+ },
1025
+ {
1026
+ "epoch": 0.94,
1027
+ "eval_loss": 1.6651471853256226,
1028
+ "eval_runtime": 37.9144,
1029
+ "eval_samples_per_second": 2.216,
1030
+ "eval_steps_per_second": 0.29,
1031
+ "step": 365
1032
+ },
1033
+ {
1034
+ "epoch": 0.95,
1035
+ "learning_rate": 5.244215938303342e-05,
1036
+ "loss": 1.6829,
1037
+ "step": 370
1038
+ },
1039
+ {
1040
+ "epoch": 0.95,
1041
+ "eval_loss": 1.6651064157485962,
1042
+ "eval_runtime": 37.7443,
1043
+ "eval_samples_per_second": 2.226,
1044
+ "eval_steps_per_second": 0.291,
1045
+ "step": 370
1046
+ },
1047
+ {
1048
+ "epoch": 0.96,
1049
+ "learning_rate": 5.1799485861182514e-05,
1050
+ "loss": 1.6923,
1051
+ "step": 375
1052
+ },
1053
+ {
1054
+ "epoch": 0.96,
1055
+ "eval_loss": 1.6646034717559814,
1056
+ "eval_runtime": 37.8783,
1057
+ "eval_samples_per_second": 2.218,
1058
+ "eval_steps_per_second": 0.29,
1059
+ "step": 375
1060
+ },
1061
+ {
1062
+ "epoch": 0.98,
1063
+ "learning_rate": 5.1156812339331615e-05,
1064
+ "loss": 1.6725,
1065
+ "step": 380
1066
+ },
1067
+ {
1068
+ "epoch": 0.98,
1069
+ "eval_loss": 1.6639235019683838,
1070
+ "eval_runtime": 37.7559,
1071
+ "eval_samples_per_second": 2.225,
1072
+ "eval_steps_per_second": 0.291,
1073
+ "step": 380
1074
+ },
1075
+ {
1076
+ "epoch": 0.99,
1077
+ "learning_rate": 5.051413881748073e-05,
1078
+ "loss": 1.6216,
1079
+ "step": 385
1080
+ },
1081
+ {
1082
+ "epoch": 0.99,
1083
+ "eval_loss": 1.6643506288528442,
1084
+ "eval_runtime": 37.9895,
1085
+ "eval_samples_per_second": 2.211,
1086
+ "eval_steps_per_second": 0.29,
1087
+ "step": 385
1088
+ },
1089
+ {
1090
+ "epoch": 1.0,
1091
+ "learning_rate": 4.987146529562982e-05,
1092
+ "loss": 1.6518,
1093
+ "step": 390
1094
+ },
1095
+ {
1096
+ "epoch": 1.0,
1097
+ "eval_loss": 1.6643046140670776,
1098
+ "eval_runtime": 37.9289,
1099
+ "eval_samples_per_second": 2.215,
1100
+ "eval_steps_per_second": 0.29,
1101
+ "step": 390
1102
+ },
1103
+ {
1104
+ "epoch": 1.02,
1105
+ "learning_rate": 4.922879177377892e-05,
1106
+ "loss": 1.658,
1107
+ "step": 395
1108
+ },
1109
+ {
1110
+ "epoch": 1.02,
1111
+ "eval_loss": 1.6641957759857178,
1112
+ "eval_runtime": 37.8638,
1113
+ "eval_samples_per_second": 2.218,
1114
+ "eval_steps_per_second": 0.291,
1115
+ "step": 395
1116
+ },
1117
+ {
1118
+ "epoch": 1.03,
1119
+ "learning_rate": 4.8586118251928024e-05,
1120
+ "loss": 1.6767,
1121
+ "step": 400
1122
+ },
1123
+ {
1124
+ "epoch": 1.03,
1125
+ "eval_loss": 1.663757085800171,
1126
+ "eval_runtime": 37.7728,
1127
+ "eval_samples_per_second": 2.224,
1128
+ "eval_steps_per_second": 0.291,
1129
+ "step": 400
1130
+ },
1131
+ {
1132
+ "epoch": 1.04,
1133
+ "learning_rate": 4.7943444730077124e-05,
1134
+ "loss": 1.6264,
1135
+ "step": 405
1136
+ },
1137
+ {
1138
+ "epoch": 1.04,
1139
+ "eval_loss": 1.6635041236877441,
1140
+ "eval_runtime": 37.7702,
1141
+ "eval_samples_per_second": 2.224,
1142
+ "eval_steps_per_second": 0.291,
1143
+ "step": 405
1144
+ },
1145
+ {
1146
+ "epoch": 1.05,
1147
+ "learning_rate": 4.7300771208226225e-05,
1148
+ "loss": 1.6527,
1149
+ "step": 410
1150
+ },
1151
+ {
1152
+ "epoch": 1.05,
1153
+ "eval_loss": 1.6632280349731445,
1154
+ "eval_runtime": 37.8453,
1155
+ "eval_samples_per_second": 2.22,
1156
+ "eval_steps_per_second": 0.291,
1157
+ "step": 410
1158
+ },
1159
+ {
1160
+ "epoch": 1.07,
1161
+ "learning_rate": 4.6658097686375325e-05,
1162
+ "loss": 1.6157,
1163
+ "step": 415
1164
+ },
1165
+ {
1166
+ "epoch": 1.07,
1167
+ "eval_loss": 1.6636598110198975,
1168
+ "eval_runtime": 37.812,
1169
+ "eval_samples_per_second": 2.222,
1170
+ "eval_steps_per_second": 0.291,
1171
+ "step": 415
1172
+ },
1173
+ {
1174
+ "epoch": 1.08,
1175
+ "learning_rate": 4.6015424164524426e-05,
1176
+ "loss": 1.5966,
1177
+ "step": 420
1178
+ },
1179
+ {
1180
+ "epoch": 1.08,
1181
+ "eval_loss": 1.663393497467041,
1182
+ "eval_runtime": 37.7743,
1183
+ "eval_samples_per_second": 2.224,
1184
+ "eval_steps_per_second": 0.291,
1185
+ "step": 420
1186
+ },
1187
+ {
1188
+ "epoch": 1.09,
1189
+ "learning_rate": 4.537275064267352e-05,
1190
+ "loss": 1.6705,
1191
+ "step": 425
1192
+ },
1193
+ {
1194
+ "epoch": 1.09,
1195
+ "eval_loss": 1.6631269454956055,
1196
+ "eval_runtime": 38.0953,
1197
+ "eval_samples_per_second": 2.205,
1198
+ "eval_steps_per_second": 0.289,
1199
+ "step": 425
1200
+ },
1201
+ {
1202
+ "epoch": 1.11,
1203
+ "learning_rate": 4.473007712082262e-05,
1204
+ "loss": 1.6691,
1205
+ "step": 430
1206
+ },
1207
+ {
1208
+ "epoch": 1.11,
1209
+ "eval_loss": 1.6633110046386719,
1210
+ "eval_runtime": 37.8964,
1211
+ "eval_samples_per_second": 2.217,
1212
+ "eval_steps_per_second": 0.29,
1213
+ "step": 430
1214
+ },
1215
+ {
1216
+ "epoch": 1.12,
1217
+ "learning_rate": 4.408740359897173e-05,
1218
+ "loss": 1.6332,
1219
+ "step": 435
1220
+ },
1221
+ {
1222
+ "epoch": 1.12,
1223
+ "eval_loss": 1.6628649234771729,
1224
+ "eval_runtime": 38.0621,
1225
+ "eval_samples_per_second": 2.207,
1226
+ "eval_steps_per_second": 0.289,
1227
+ "step": 435
1228
+ },
1229
+ {
1230
+ "epoch": 1.13,
1231
+ "learning_rate": 4.344473007712083e-05,
1232
+ "loss": 1.5916,
1233
+ "step": 440
1234
+ },
1235
+ {
1236
+ "epoch": 1.13,
1237
+ "eval_loss": 1.662834882736206,
1238
+ "eval_runtime": 37.9722,
1239
+ "eval_samples_per_second": 2.212,
1240
+ "eval_steps_per_second": 0.29,
1241
+ "step": 440
1242
+ },
1243
+ {
1244
+ "epoch": 1.14,
1245
+ "learning_rate": 4.280205655526993e-05,
1246
+ "loss": 1.6543,
1247
+ "step": 445
1248
+ },
1249
+ {
1250
+ "epoch": 1.14,
1251
+ "eval_loss": 1.6630265712738037,
1252
+ "eval_runtime": 37.9699,
1253
+ "eval_samples_per_second": 2.212,
1254
+ "eval_steps_per_second": 0.29,
1255
+ "step": 445
1256
+ },
1257
+ {
1258
+ "epoch": 1.16,
1259
+ "learning_rate": 4.215938303341902e-05,
1260
+ "loss": 1.6353,
1261
+ "step": 450
1262
+ },
1263
+ {
1264
+ "epoch": 1.16,
1265
+ "eval_loss": 1.6625033617019653,
1266
+ "eval_runtime": 37.7505,
1267
+ "eval_samples_per_second": 2.225,
1268
+ "eval_steps_per_second": 0.291,
1269
+ "step": 450
1270
+ },
1271
+ {
1272
+ "epoch": 1.17,
1273
+ "learning_rate": 4.151670951156812e-05,
1274
+ "loss": 1.6441,
1275
+ "step": 455
1276
+ },
1277
+ {
1278
+ "epoch": 1.17,
1279
+ "eval_loss": 1.662785530090332,
1280
+ "eval_runtime": 37.75,
1281
+ "eval_samples_per_second": 2.225,
1282
+ "eval_steps_per_second": 0.291,
1283
+ "step": 455
1284
+ },
1285
+ {
1286
+ "epoch": 1.18,
1287
+ "learning_rate": 4.0874035989717224e-05,
1288
+ "loss": 1.6631,
1289
+ "step": 460
1290
+ },
1291
+ {
1292
+ "epoch": 1.18,
1293
+ "eval_loss": 1.6627779006958008,
1294
+ "eval_runtime": 37.864,
1295
+ "eval_samples_per_second": 2.218,
1296
+ "eval_steps_per_second": 0.291,
1297
+ "step": 460
1298
+ },
1299
+ {
1300
+ "epoch": 1.2,
1301
+ "learning_rate": 4.0231362467866324e-05,
1302
+ "loss": 1.6327,
1303
+ "step": 465
1304
+ },
1305
+ {
1306
+ "epoch": 1.2,
1307
+ "eval_loss": 1.662458896636963,
1308
+ "eval_runtime": 37.7496,
1309
+ "eval_samples_per_second": 2.225,
1310
+ "eval_steps_per_second": 0.291,
1311
+ "step": 465
1312
+ },
1313
+ {
1314
+ "epoch": 1.21,
1315
+ "learning_rate": 3.958868894601543e-05,
1316
+ "loss": 1.6238,
1317
+ "step": 470
1318
+ },
1319
+ {
1320
+ "epoch": 1.21,
1321
+ "eval_loss": 1.6621876955032349,
1322
+ "eval_runtime": 37.8943,
1323
+ "eval_samples_per_second": 2.217,
1324
+ "eval_steps_per_second": 0.29,
1325
+ "step": 470
1326
+ },
1327
+ {
1328
+ "epoch": 1.22,
1329
+ "learning_rate": 3.8946015424164526e-05,
1330
+ "loss": 1.6231,
1331
+ "step": 475
1332
+ },
1333
+ {
1334
+ "epoch": 1.22,
1335
+ "eval_loss": 1.662013292312622,
1336
+ "eval_runtime": 37.9364,
1337
+ "eval_samples_per_second": 2.214,
1338
+ "eval_steps_per_second": 0.29,
1339
+ "step": 475
1340
+ },
1341
+ {
1342
+ "epoch": 1.23,
1343
+ "learning_rate": 3.8303341902313626e-05,
1344
+ "loss": 1.6381,
1345
+ "step": 480
1346
+ },
1347
+ {
1348
+ "epoch": 1.23,
1349
+ "eval_loss": 1.6615785360336304,
1350
+ "eval_runtime": 37.91,
1351
+ "eval_samples_per_second": 2.216,
1352
+ "eval_steps_per_second": 0.29,
1353
+ "step": 480
1354
+ },
1355
+ {
1356
+ "epoch": 1.25,
1357
+ "learning_rate": 3.766066838046273e-05,
1358
+ "loss": 1.6419,
1359
+ "step": 485
1360
+ },
1361
+ {
1362
+ "epoch": 1.25,
1363
+ "eval_loss": 1.662042498588562,
1364
+ "eval_runtime": 37.7834,
1365
+ "eval_samples_per_second": 2.223,
1366
+ "eval_steps_per_second": 0.291,
1367
+ "step": 485
1368
+ },
1369
+ {
1370
+ "epoch": 1.26,
1371
+ "learning_rate": 3.701799485861183e-05,
1372
+ "loss": 1.6319,
1373
+ "step": 490
1374
+ },
1375
+ {
1376
+ "epoch": 1.26,
1377
+ "eval_loss": 1.6619195938110352,
1378
+ "eval_runtime": 37.7385,
1379
+ "eval_samples_per_second": 2.226,
1380
+ "eval_steps_per_second": 0.291,
1381
+ "step": 490
1382
+ },
1383
+ {
1384
+ "epoch": 1.27,
1385
+ "learning_rate": 3.637532133676093e-05,
1386
+ "loss": 1.609,
1387
+ "step": 495
1388
+ },
1389
+ {
1390
+ "epoch": 1.27,
1391
+ "eval_loss": 1.6615225076675415,
1392
+ "eval_runtime": 37.8056,
1393
+ "eval_samples_per_second": 2.222,
1394
+ "eval_steps_per_second": 0.291,
1395
+ "step": 495
1396
+ },
1397
+ {
1398
+ "epoch": 1.29,
1399
+ "learning_rate": 3.573264781491003e-05,
1400
+ "loss": 1.6435,
1401
+ "step": 500
1402
+ },
1403
+ {
1404
+ "epoch": 1.29,
1405
+ "eval_loss": 1.6610548496246338,
1406
+ "eval_runtime": 37.9353,
1407
+ "eval_samples_per_second": 2.214,
1408
+ "eval_steps_per_second": 0.29,
1409
+ "step": 500
1410
+ },
1411
+ {
1412
+ "epoch": 1.3,
1413
+ "learning_rate": 3.508997429305913e-05,
1414
+ "loss": 1.6381,
1415
+ "step": 505
1416
+ },
1417
+ {
1418
+ "epoch": 1.3,
1419
+ "eval_loss": 1.661049246788025,
1420
+ "eval_runtime": 37.9669,
1421
+ "eval_samples_per_second": 2.212,
1422
+ "eval_steps_per_second": 0.29,
1423
+ "step": 505
1424
+ },
1425
+ {
1426
+ "epoch": 1.31,
1427
+ "learning_rate": 3.444730077120823e-05,
1428
+ "loss": 1.6506,
1429
+ "step": 510
1430
+ },
1431
+ {
1432
+ "epoch": 1.31,
1433
+ "eval_loss": 1.6610089540481567,
1434
+ "eval_runtime": 37.9127,
1435
+ "eval_samples_per_second": 2.216,
1436
+ "eval_steps_per_second": 0.29,
1437
+ "step": 510
1438
+ },
1439
+ {
1440
+ "epoch": 1.32,
1441
+ "learning_rate": 3.380462724935733e-05,
1442
+ "loss": 1.6376,
1443
+ "step": 515
1444
+ },
1445
+ {
1446
+ "epoch": 1.32,
1447
+ "eval_loss": 1.6608952283859253,
1448
+ "eval_runtime": 37.9454,
1449
+ "eval_samples_per_second": 2.214,
1450
+ "eval_steps_per_second": 0.29,
1451
+ "step": 515
1452
+ },
1453
+ {
1454
+ "epoch": 1.34,
1455
+ "learning_rate": 3.316195372750643e-05,
1456
+ "loss": 1.6354,
1457
+ "step": 520
1458
+ },
1459
+ {
1460
+ "epoch": 1.34,
1461
+ "eval_loss": 1.6610838174819946,
1462
+ "eval_runtime": 37.7265,
1463
+ "eval_samples_per_second": 2.227,
1464
+ "eval_steps_per_second": 0.292,
1465
+ "step": 520
1466
+ },
1467
+ {
1468
+ "epoch": 1.35,
1469
+ "learning_rate": 3.251928020565553e-05,
1470
+ "loss": 1.6201,
1471
+ "step": 525
1472
+ },
1473
+ {
1474
+ "epoch": 1.35,
1475
+ "eval_loss": 1.6610314846038818,
1476
+ "eval_runtime": 37.8464,
1477
+ "eval_samples_per_second": 2.219,
1478
+ "eval_steps_per_second": 0.291,
1479
+ "step": 525
1480
+ },
1481
+ {
1482
+ "epoch": 1.36,
1483
+ "learning_rate": 3.1876606683804625e-05,
1484
+ "loss": 1.6461,
1485
+ "step": 530
1486
+ },
1487
+ {
1488
+ "epoch": 1.36,
1489
+ "eval_loss": 1.6605802774429321,
1490
+ "eval_runtime": 37.9503,
1491
+ "eval_samples_per_second": 2.213,
1492
+ "eval_steps_per_second": 0.29,
1493
+ "step": 530
1494
+ },
1495
+ {
1496
+ "epoch": 1.38,
1497
+ "learning_rate": 3.1233933161953726e-05,
1498
+ "loss": 1.6818,
1499
+ "step": 535
1500
+ },
1501
+ {
1502
+ "epoch": 1.38,
1503
+ "eval_loss": 1.6607571840286255,
1504
+ "eval_runtime": 37.9086,
1505
+ "eval_samples_per_second": 2.216,
1506
+ "eval_steps_per_second": 0.29,
1507
+ "step": 535
1508
+ },
1509
+ {
1510
+ "epoch": 1.39,
1511
+ "learning_rate": 3.059125964010283e-05,
1512
+ "loss": 1.6117,
1513
+ "step": 540
1514
+ },
1515
+ {
1516
+ "epoch": 1.39,
1517
+ "eval_loss": 1.6605336666107178,
1518
+ "eval_runtime": 37.947,
1519
+ "eval_samples_per_second": 2.214,
1520
+ "eval_steps_per_second": 0.29,
1521
+ "step": 540
1522
+ },
1523
+ {
1524
+ "epoch": 1.4,
1525
+ "learning_rate": 2.994858611825193e-05,
1526
+ "loss": 1.6252,
1527
+ "step": 545
1528
+ },
1529
+ {
1530
+ "epoch": 1.4,
1531
+ "eval_loss": 1.6603213548660278,
1532
+ "eval_runtime": 37.9826,
1533
+ "eval_samples_per_second": 2.212,
1534
+ "eval_steps_per_second": 0.29,
1535
+ "step": 545
1536
+ },
1537
+ {
1538
+ "epoch": 1.41,
1539
+ "learning_rate": 2.930591259640103e-05,
1540
+ "loss": 1.6486,
1541
+ "step": 550
1542
+ },
1543
+ {
1544
+ "epoch": 1.41,
1545
+ "eval_loss": 1.6600297689437866,
1546
+ "eval_runtime": 37.9635,
1547
+ "eval_samples_per_second": 2.213,
1548
+ "eval_steps_per_second": 0.29,
1549
+ "step": 550
1550
+ },
1551
+ {
1552
+ "epoch": 1.43,
1553
+ "learning_rate": 2.866323907455013e-05,
1554
+ "loss": 1.6861,
1555
+ "step": 555
1556
+ },
1557
+ {
1558
+ "epoch": 1.43,
1559
+ "eval_loss": 1.6602742671966553,
1560
+ "eval_runtime": 37.8722,
1561
+ "eval_samples_per_second": 2.218,
1562
+ "eval_steps_per_second": 0.29,
1563
+ "step": 555
1564
+ },
1565
+ {
1566
+ "epoch": 1.44,
1567
+ "learning_rate": 2.802056555269923e-05,
1568
+ "loss": 1.6624,
1569
+ "step": 560
1570
+ },
1571
+ {
1572
+ "epoch": 1.44,
1573
+ "eval_loss": 1.6599992513656616,
1574
+ "eval_runtime": 37.9306,
1575
+ "eval_samples_per_second": 2.215,
1576
+ "eval_steps_per_second": 0.29,
1577
+ "step": 560
1578
+ },
1579
+ {
1580
+ "epoch": 1.45,
1581
+ "learning_rate": 2.737789203084833e-05,
1582
+ "loss": 1.6323,
1583
+ "step": 565
1584
+ },
1585
+ {
1586
+ "epoch": 1.45,
1587
+ "eval_loss": 1.6600306034088135,
1588
+ "eval_runtime": 37.9548,
1589
+ "eval_samples_per_second": 2.213,
1590
+ "eval_steps_per_second": 0.29,
1591
+ "step": 565
1592
+ },
1593
+ {
1594
+ "epoch": 1.47,
1595
+ "learning_rate": 2.673521850899743e-05,
1596
+ "loss": 1.6707,
1597
+ "step": 570
1598
+ },
1599
+ {
1600
+ "epoch": 1.47,
1601
+ "eval_loss": 1.6601014137268066,
1602
+ "eval_runtime": 38.0349,
1603
+ "eval_samples_per_second": 2.208,
1604
+ "eval_steps_per_second": 0.289,
1605
+ "step": 570
1606
+ },
1607
+ {
1608
+ "epoch": 1.48,
1609
+ "learning_rate": 2.6092544987146534e-05,
1610
+ "loss": 1.6478,
1611
+ "step": 575
1612
+ },
1613
+ {
1614
+ "epoch": 1.48,
1615
+ "eval_loss": 1.6597967147827148,
1616
+ "eval_runtime": 37.9413,
1617
+ "eval_samples_per_second": 2.214,
1618
+ "eval_steps_per_second": 0.29,
1619
+ "step": 575
1620
+ },
1621
+ {
1622
+ "epoch": 1.49,
1623
+ "learning_rate": 2.5449871465295634e-05,
1624
+ "loss": 1.6715,
1625
+ "step": 580
1626
+ },
1627
+ {
1628
+ "epoch": 1.49,
1629
+ "eval_loss": 1.6598337888717651,
1630
+ "eval_runtime": 37.8386,
1631
+ "eval_samples_per_second": 2.22,
1632
+ "eval_steps_per_second": 0.291,
1633
+ "step": 580
1634
+ },
1635
+ {
1636
+ "epoch": 1.5,
1637
+ "learning_rate": 2.480719794344473e-05,
1638
+ "loss": 1.6626,
1639
+ "step": 585
1640
+ },
1641
+ {
1642
+ "epoch": 1.5,
1643
+ "eval_loss": 1.6600011587142944,
1644
+ "eval_runtime": 37.7901,
1645
+ "eval_samples_per_second": 2.223,
1646
+ "eval_steps_per_second": 0.291,
1647
+ "step": 585
1648
+ },
1649
+ {
1650
+ "epoch": 1.52,
1651
+ "learning_rate": 2.4164524421593832e-05,
1652
+ "loss": 1.651,
1653
+ "step": 590
1654
+ },
1655
+ {
1656
+ "epoch": 1.52,
1657
+ "eval_loss": 1.65969717502594,
1658
+ "eval_runtime": 37.7643,
1659
+ "eval_samples_per_second": 2.224,
1660
+ "eval_steps_per_second": 0.291,
1661
+ "step": 590
1662
+ },
1663
+ {
1664
+ "epoch": 1.53,
1665
+ "learning_rate": 2.3521850899742933e-05,
1666
+ "loss": 1.6346,
1667
+ "step": 595
1668
+ },
1669
+ {
1670
+ "epoch": 1.53,
1671
+ "eval_loss": 1.6598676443099976,
1672
+ "eval_runtime": 37.9026,
1673
+ "eval_samples_per_second": 2.216,
1674
+ "eval_steps_per_second": 0.29,
1675
+ "step": 595
1676
+ },
1677
+ {
1678
+ "epoch": 1.54,
1679
+ "learning_rate": 2.2879177377892033e-05,
1680
+ "loss": 1.6533,
1681
+ "step": 600
1682
+ },
1683
+ {
1684
+ "epoch": 1.54,
1685
+ "eval_loss": 1.659874677658081,
1686
+ "eval_runtime": 37.9871,
1687
+ "eval_samples_per_second": 2.211,
1688
+ "eval_steps_per_second": 0.29,
1689
+ "step": 600
1690
+ },
1691
+ {
1692
+ "epoch": 1.56,
1693
+ "learning_rate": 2.2236503856041134e-05,
1694
+ "loss": 1.6913,
1695
+ "step": 605
1696
+ },
1697
+ {
1698
+ "epoch": 1.56,
1699
+ "eval_loss": 1.659849762916565,
1700
+ "eval_runtime": 37.8712,
1701
+ "eval_samples_per_second": 2.218,
1702
+ "eval_steps_per_second": 0.29,
1703
+ "step": 605
1704
+ },
1705
+ {
1706
+ "epoch": 1.57,
1707
+ "learning_rate": 2.159383033419023e-05,
1708
+ "loss": 1.6709,
1709
+ "step": 610
1710
+ },
1711
+ {
1712
+ "epoch": 1.57,
1713
+ "eval_loss": 1.6598856449127197,
1714
+ "eval_runtime": 37.7043,
1715
+ "eval_samples_per_second": 2.228,
1716
+ "eval_steps_per_second": 0.292,
1717
+ "step": 610
1718
+ },
1719
+ {
1720
+ "epoch": 1.58,
1721
+ "learning_rate": 2.095115681233933e-05,
1722
+ "loss": 1.598,
1723
+ "step": 615
1724
+ },
1725
+ {
1726
+ "epoch": 1.58,
1727
+ "eval_loss": 1.6597983837127686,
1728
+ "eval_runtime": 37.722,
1729
+ "eval_samples_per_second": 2.227,
1730
+ "eval_steps_per_second": 0.292,
1731
+ "step": 615
1732
+ },
1733
+ {
1734
+ "epoch": 1.59,
1735
+ "learning_rate": 2.0308483290488432e-05,
1736
+ "loss": 1.6434,
1737
+ "step": 620
1738
+ },
1739
+ {
1740
+ "epoch": 1.59,
1741
+ "eval_loss": 1.6595433950424194,
1742
+ "eval_runtime": 37.6125,
1743
+ "eval_samples_per_second": 2.233,
1744
+ "eval_steps_per_second": 0.292,
1745
+ "step": 620
1746
+ },
1747
+ {
1748
+ "epoch": 1.61,
1749
+ "learning_rate": 1.9665809768637533e-05,
1750
+ "loss": 1.6086,
1751
+ "step": 625
1752
+ },
1753
+ {
1754
+ "epoch": 1.61,
1755
+ "eval_loss": 1.6594945192337036,
1756
+ "eval_runtime": 38.0046,
1757
+ "eval_samples_per_second": 2.21,
1758
+ "eval_steps_per_second": 0.289,
1759
+ "step": 625
1760
+ },
1761
+ {
1762
+ "epoch": 1.62,
1763
+ "learning_rate": 1.9023136246786633e-05,
1764
+ "loss": 1.5962,
1765
+ "step": 630
1766
+ },
1767
+ {
1768
+ "epoch": 1.62,
1769
+ "eval_loss": 1.6595605611801147,
1770
+ "eval_runtime": 37.8518,
1771
+ "eval_samples_per_second": 2.219,
1772
+ "eval_steps_per_second": 0.291,
1773
+ "step": 630
1774
+ },
1775
+ {
1776
+ "epoch": 1.63,
1777
+ "learning_rate": 1.8380462724935734e-05,
1778
+ "loss": 1.6695,
1779
+ "step": 635
1780
+ },
1781
+ {
1782
+ "epoch": 1.63,
1783
+ "eval_loss": 1.6592403650283813,
1784
+ "eval_runtime": 37.7285,
1785
+ "eval_samples_per_second": 2.226,
1786
+ "eval_steps_per_second": 0.292,
1787
+ "step": 635
1788
+ },
1789
+ {
1790
+ "epoch": 1.65,
1791
+ "learning_rate": 1.7737789203084834e-05,
1792
+ "loss": 1.6526,
1793
+ "step": 640
1794
+ },
1795
+ {
1796
+ "epoch": 1.65,
1797
+ "eval_loss": 1.6591880321502686,
1798
+ "eval_runtime": 37.758,
1799
+ "eval_samples_per_second": 2.225,
1800
+ "eval_steps_per_second": 0.291,
1801
+ "step": 640
1802
+ },
1803
+ {
1804
+ "epoch": 1.66,
1805
+ "learning_rate": 1.7095115681233935e-05,
1806
+ "loss": 1.672,
1807
+ "step": 645
1808
+ },
1809
+ {
1810
+ "epoch": 1.66,
1811
+ "eval_loss": 1.6588138341903687,
1812
+ "eval_runtime": 37.92,
1813
+ "eval_samples_per_second": 2.215,
1814
+ "eval_steps_per_second": 0.29,
1815
+ "step": 645
1816
+ },
1817
+ {
1818
+ "epoch": 1.67,
1819
+ "learning_rate": 1.6452442159383032e-05,
1820
+ "loss": 1.6347,
1821
+ "step": 650
1822
+ },
1823
+ {
1824
+ "epoch": 1.67,
1825
+ "eval_loss": 1.6589809656143188,
1826
+ "eval_runtime": 37.8021,
1827
+ "eval_samples_per_second": 2.222,
1828
+ "eval_steps_per_second": 0.291,
1829
+ "step": 650
1830
+ },
1831
+ {
1832
+ "epoch": 1.68,
1833
+ "learning_rate": 1.5809768637532136e-05,
1834
+ "loss": 1.6227,
1835
+ "step": 655
1836
+ },
1837
+ {
1838
+ "epoch": 1.68,
1839
+ "eval_loss": 1.6586792469024658,
1840
+ "eval_runtime": 37.9128,
1841
+ "eval_samples_per_second": 2.216,
1842
+ "eval_steps_per_second": 0.29,
1843
+ "step": 655
1844
+ },
1845
+ {
1846
+ "epoch": 1.7,
1847
+ "learning_rate": 1.5167095115681235e-05,
1848
+ "loss": 1.6308,
1849
+ "step": 660
1850
+ },
1851
+ {
1852
+ "epoch": 1.7,
1853
+ "eval_loss": 1.6585420370101929,
1854
+ "eval_runtime": 37.9881,
1855
+ "eval_samples_per_second": 2.211,
1856
+ "eval_steps_per_second": 0.29,
1857
+ "step": 660
1858
+ },
1859
+ {
1860
+ "epoch": 1.71,
1861
+ "learning_rate": 1.4524421593830334e-05,
1862
+ "loss": 1.6495,
1863
+ "step": 665
1864
+ },
1865
+ {
1866
+ "epoch": 1.71,
1867
+ "eval_loss": 1.6587530374526978,
1868
+ "eval_runtime": 37.9148,
1869
+ "eval_samples_per_second": 2.215,
1870
+ "eval_steps_per_second": 0.29,
1871
+ "step": 665
1872
+ },
1873
+ {
1874
+ "epoch": 1.72,
1875
+ "learning_rate": 1.3881748071979436e-05,
1876
+ "loss": 1.6712,
1877
+ "step": 670
1878
+ },
1879
+ {
1880
+ "epoch": 1.72,
1881
+ "eval_loss": 1.658648133277893,
1882
+ "eval_runtime": 37.9313,
1883
+ "eval_samples_per_second": 2.215,
1884
+ "eval_steps_per_second": 0.29,
1885
+ "step": 670
1886
+ },
1887
+ {
1888
+ "epoch": 1.74,
1889
+ "learning_rate": 1.3239074550128535e-05,
1890
+ "loss": 1.636,
1891
+ "step": 675
1892
+ },
1893
+ {
1894
+ "epoch": 1.74,
1895
+ "eval_loss": 1.6584407091140747,
1896
+ "eval_runtime": 37.7963,
1897
+ "eval_samples_per_second": 2.222,
1898
+ "eval_steps_per_second": 0.291,
1899
+ "step": 675
1900
+ },
1901
+ {
1902
+ "epoch": 1.75,
1903
+ "learning_rate": 1.2596401028277636e-05,
1904
+ "loss": 1.6453,
1905
+ "step": 680
1906
+ },
1907
+ {
1908
+ "epoch": 1.75,
1909
+ "eval_loss": 1.658594012260437,
1910
+ "eval_runtime": 37.7226,
1911
+ "eval_samples_per_second": 2.227,
1912
+ "eval_steps_per_second": 0.292,
1913
+ "step": 680
1914
+ },
1915
+ {
1916
+ "epoch": 1.76,
1917
+ "learning_rate": 1.1953727506426736e-05,
1918
+ "loss": 1.6509,
1919
+ "step": 685
1920
+ },
1921
+ {
1922
+ "epoch": 1.76,
1923
+ "eval_loss": 1.658467411994934,
1924
+ "eval_runtime": 37.8179,
1925
+ "eval_samples_per_second": 2.221,
1926
+ "eval_steps_per_second": 0.291,
1927
+ "step": 685
1928
+ },
1929
+ {
1930
+ "epoch": 1.77,
1931
+ "learning_rate": 1.1311053984575835e-05,
1932
+ "loss": 1.6489,
1933
+ "step": 690
1934
+ },
1935
+ {
1936
+ "epoch": 1.77,
1937
+ "eval_loss": 1.6585352420806885,
1938
+ "eval_runtime": 37.6768,
1939
+ "eval_samples_per_second": 2.229,
1940
+ "eval_steps_per_second": 0.292,
1941
+ "step": 690
1942
+ },
1943
+ {
1944
+ "epoch": 1.79,
1945
+ "learning_rate": 1.0668380462724936e-05,
1946
+ "loss": 1.6424,
1947
+ "step": 695
1948
+ },
1949
+ {
1950
+ "epoch": 1.79,
1951
+ "eval_loss": 1.6582835912704468,
1952
+ "eval_runtime": 38.0083,
1953
+ "eval_samples_per_second": 2.21,
1954
+ "eval_steps_per_second": 0.289,
1955
+ "step": 695
1956
+ },
1957
+ {
1958
+ "epoch": 1.8,
1959
+ "learning_rate": 1.0025706940874038e-05,
1960
+ "loss": 1.6131,
1961
+ "step": 700
1962
+ },
1963
+ {
1964
+ "epoch": 1.8,
1965
+ "eval_loss": 1.658105731010437,
1966
+ "eval_runtime": 37.8711,
1967
+ "eval_samples_per_second": 2.218,
1968
+ "eval_steps_per_second": 0.29,
1969
+ "step": 700
1970
+ },
1971
+ {
1972
+ "epoch": 1.81,
1973
+ "learning_rate": 9.383033419023137e-06,
1974
+ "loss": 1.6225,
1975
+ "step": 705
1976
+ },
1977
+ {
1978
+ "epoch": 1.81,
1979
+ "eval_loss": 1.658282995223999,
1980
+ "eval_runtime": 37.8125,
1981
+ "eval_samples_per_second": 2.221,
1982
+ "eval_steps_per_second": 0.291,
1983
+ "step": 705
1984
+ },
1985
+ {
1986
+ "epoch": 1.83,
1987
+ "learning_rate": 8.740359897172237e-06,
1988
+ "loss": 1.6928,
1989
+ "step": 710
1990
+ },
1991
+ {
1992
+ "epoch": 1.83,
1993
+ "eval_loss": 1.6581056118011475,
1994
+ "eval_runtime": 37.863,
1995
+ "eval_samples_per_second": 2.219,
1996
+ "eval_steps_per_second": 0.291,
1997
+ "step": 710
1998
+ },
1999
+ {
2000
+ "epoch": 1.84,
2001
+ "learning_rate": 8.097686375321336e-06,
2002
+ "loss": 1.6227,
2003
+ "step": 715
2004
+ },
2005
+ {
2006
+ "epoch": 1.84,
2007
+ "eval_loss": 1.6581188440322876,
2008
+ "eval_runtime": 37.8646,
2009
+ "eval_samples_per_second": 2.218,
2010
+ "eval_steps_per_second": 0.291,
2011
+ "step": 715
2012
+ },
2013
+ {
2014
+ "epoch": 1.85,
2015
+ "learning_rate": 7.4550128534704376e-06,
2016
+ "loss": 1.6593,
2017
+ "step": 720
2018
+ },
2019
+ {
2020
+ "epoch": 1.85,
2021
+ "eval_loss": 1.6579999923706055,
2022
+ "eval_runtime": 37.876,
2023
+ "eval_samples_per_second": 2.218,
2024
+ "eval_steps_per_second": 0.29,
2025
+ "step": 720
2026
+ },
2027
+ {
2028
+ "epoch": 1.86,
2029
+ "learning_rate": 6.812339331619537e-06,
2030
+ "loss": 1.6642,
2031
+ "step": 725
2032
+ },
2033
+ {
2034
+ "epoch": 1.86,
2035
+ "eval_loss": 1.65813410282135,
2036
+ "eval_runtime": 37.7093,
2037
+ "eval_samples_per_second": 2.228,
2038
+ "eval_steps_per_second": 0.292,
2039
+ "step": 725
2040
+ },
2041
+ {
2042
+ "epoch": 1.88,
2043
+ "learning_rate": 6.169665809768638e-06,
2044
+ "loss": 1.6414,
2045
+ "step": 730
2046
+ },
2047
+ {
2048
+ "epoch": 1.88,
2049
+ "eval_loss": 1.6581989526748657,
2050
+ "eval_runtime": 37.7818,
2051
+ "eval_samples_per_second": 2.223,
2052
+ "eval_steps_per_second": 0.291,
2053
+ "step": 730
2054
+ },
2055
+ {
2056
+ "epoch": 1.89,
2057
+ "learning_rate": 5.526992287917738e-06,
2058
+ "loss": 1.6519,
2059
+ "step": 735
2060
+ },
2061
+ {
2062
+ "epoch": 1.89,
2063
+ "eval_loss": 1.6577394008636475,
2064
+ "eval_runtime": 37.7264,
2065
+ "eval_samples_per_second": 2.227,
2066
+ "eval_steps_per_second": 0.292,
2067
+ "step": 735
2068
+ },
2069
+ {
2070
+ "epoch": 1.9,
2071
+ "learning_rate": 4.884318766066838e-06,
2072
+ "loss": 1.6485,
2073
+ "step": 740
2074
+ },
2075
+ {
2076
+ "epoch": 1.9,
2077
+ "eval_loss": 1.6580238342285156,
2078
+ "eval_runtime": 37.8729,
2079
+ "eval_samples_per_second": 2.218,
2080
+ "eval_steps_per_second": 0.29,
2081
+ "step": 740
2082
+ },
2083
+ {
2084
+ "epoch": 1.92,
2085
+ "learning_rate": 4.241645244215939e-06,
2086
+ "loss": 1.6261,
2087
+ "step": 745
2088
+ },
2089
+ {
2090
+ "epoch": 1.92,
2091
+ "eval_loss": 1.6580148935317993,
2092
+ "eval_runtime": 37.9546,
2093
+ "eval_samples_per_second": 2.213,
2094
+ "eval_steps_per_second": 0.29,
2095
+ "step": 745
2096
+ },
2097
+ {
2098
+ "epoch": 1.93,
2099
+ "learning_rate": 3.598971722365039e-06,
2100
+ "loss": 1.657,
2101
+ "step": 750
2102
+ },
2103
+ {
2104
+ "epoch": 1.93,
2105
+ "eval_loss": 1.657867670059204,
2106
+ "eval_runtime": 37.9411,
2107
+ "eval_samples_per_second": 2.214,
2108
+ "eval_steps_per_second": 0.29,
2109
+ "step": 750
2110
+ },
2111
+ {
2112
+ "epoch": 1.94,
2113
+ "learning_rate": 2.956298200514139e-06,
2114
+ "loss": 1.6535,
2115
+ "step": 755
2116
+ },
2117
+ {
2118
+ "epoch": 1.94,
2119
+ "eval_loss": 1.6581624746322632,
2120
+ "eval_runtime": 37.8532,
2121
+ "eval_samples_per_second": 2.219,
2122
+ "eval_steps_per_second": 0.291,
2123
+ "step": 755
2124
+ },
2125
+ {
2126
+ "epoch": 1.95,
2127
+ "learning_rate": 2.313624678663239e-06,
2128
+ "loss": 1.6518,
2129
+ "step": 760
2130
+ },
2131
+ {
2132
+ "epoch": 1.95,
2133
+ "eval_loss": 1.6577980518341064,
2134
+ "eval_runtime": 37.657,
2135
+ "eval_samples_per_second": 2.231,
2136
+ "eval_steps_per_second": 0.292,
2137
+ "step": 760
2138
+ }
2139
+ ],
2140
+ "logging_steps": 5,
2141
+ "max_steps": 778,
2142
+ "num_train_epochs": 2,
2143
+ "save_steps": 10,
2144
+ "total_flos": 2.4689619442335744e+17,
2145
+ "trial_name": null,
2146
+ "trial_params": null
2147
+ }
checkpoint-760/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6211dd8a7344cda681d69cc75786fde8b7049338b76992777b83de12eab260a
3
+ size 4027
checkpoint-770/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: True
10
+ - load_in_4bit: False
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: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.5.0
checkpoint-770/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
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": 32,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-770/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b23a07dab07f00f90900bcaff0b20996d79c23889b719aa84f8ba4c9a38f6ce9
3
+ size 16822989
checkpoint-770/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0121c2817333bbb04bb24f3fb1e93c76c739f0a35507d060bc303a5685e3b26
3
+ size 33665221
checkpoint-770/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce69322a0e59f5f685c2994b6fbecb094be404ff54b66822e0afe5f8f931c843
3
+ size 14575
checkpoint-770/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e0045306b6a056511f2064b926761c66585bc387da6ef2b13833abefc9995ed
3
+ size 627
checkpoint-770/trainer_state.json ADDED
@@ -0,0 +1,2175 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.6577980518341064,
3
+ "best_model_checkpoint": "/scratch/kwamea/llama-output/checkpoint-760",
4
+ "epoch": 1.9794344473007712,
5
+ "eval_steps": 5,
6
+ "global_step": 770,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 9.93573264781491e-05,
14
+ "loss": 1.9824,
15
+ "step": 5
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "eval_loss": 1.954480528831482,
20
+ "eval_runtime": 38.0417,
21
+ "eval_samples_per_second": 2.208,
22
+ "eval_steps_per_second": 0.289,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.03,
27
+ "learning_rate": 9.87146529562982e-05,
28
+ "loss": 1.9245,
29
+ "step": 10
30
+ },
31
+ {
32
+ "epoch": 0.03,
33
+ "eval_loss": 1.8942010402679443,
34
+ "eval_runtime": 37.8934,
35
+ "eval_samples_per_second": 2.217,
36
+ "eval_steps_per_second": 0.29,
37
+ "step": 10
38
+ },
39
+ {
40
+ "epoch": 0.04,
41
+ "learning_rate": 9.80719794344473e-05,
42
+ "loss": 1.8616,
43
+ "step": 15
44
+ },
45
+ {
46
+ "epoch": 0.04,
47
+ "eval_loss": 1.8424055576324463,
48
+ "eval_runtime": 37.8727,
49
+ "eval_samples_per_second": 2.218,
50
+ "eval_steps_per_second": 0.29,
51
+ "step": 15
52
+ },
53
+ {
54
+ "epoch": 0.05,
55
+ "learning_rate": 9.742930591259641e-05,
56
+ "loss": 1.8267,
57
+ "step": 20
58
+ },
59
+ {
60
+ "epoch": 0.05,
61
+ "eval_loss": 1.8279685974121094,
62
+ "eval_runtime": 38.166,
63
+ "eval_samples_per_second": 2.201,
64
+ "eval_steps_per_second": 0.288,
65
+ "step": 20
66
+ },
67
+ {
68
+ "epoch": 0.06,
69
+ "learning_rate": 9.67866323907455e-05,
70
+ "loss": 1.835,
71
+ "step": 25
72
+ },
73
+ {
74
+ "epoch": 0.06,
75
+ "eval_loss": 1.8134318590164185,
76
+ "eval_runtime": 38.0411,
77
+ "eval_samples_per_second": 2.208,
78
+ "eval_steps_per_second": 0.289,
79
+ "step": 25
80
+ },
81
+ {
82
+ "epoch": 0.08,
83
+ "learning_rate": 9.61439588688946e-05,
84
+ "loss": 1.8236,
85
+ "step": 30
86
+ },
87
+ {
88
+ "epoch": 0.08,
89
+ "eval_loss": 1.8010995388031006,
90
+ "eval_runtime": 37.8753,
91
+ "eval_samples_per_second": 2.218,
92
+ "eval_steps_per_second": 0.29,
93
+ "step": 30
94
+ },
95
+ {
96
+ "epoch": 0.09,
97
+ "learning_rate": 9.550128534704372e-05,
98
+ "loss": 1.7186,
99
+ "step": 35
100
+ },
101
+ {
102
+ "epoch": 0.09,
103
+ "eval_loss": 1.7901933193206787,
104
+ "eval_runtime": 38.0374,
105
+ "eval_samples_per_second": 2.208,
106
+ "eval_steps_per_second": 0.289,
107
+ "step": 35
108
+ },
109
+ {
110
+ "epoch": 0.1,
111
+ "learning_rate": 9.485861182519281e-05,
112
+ "loss": 1.7959,
113
+ "step": 40
114
+ },
115
+ {
116
+ "epoch": 0.1,
117
+ "eval_loss": 1.7807564735412598,
118
+ "eval_runtime": 37.8784,
119
+ "eval_samples_per_second": 2.218,
120
+ "eval_steps_per_second": 0.29,
121
+ "step": 40
122
+ },
123
+ {
124
+ "epoch": 0.12,
125
+ "learning_rate": 9.421593830334192e-05,
126
+ "loss": 1.715,
127
+ "step": 45
128
+ },
129
+ {
130
+ "epoch": 0.12,
131
+ "eval_loss": 1.7709146738052368,
132
+ "eval_runtime": 37.8705,
133
+ "eval_samples_per_second": 2.218,
134
+ "eval_steps_per_second": 0.29,
135
+ "step": 45
136
+ },
137
+ {
138
+ "epoch": 0.13,
139
+ "learning_rate": 9.357326478149101e-05,
140
+ "loss": 1.7581,
141
+ "step": 50
142
+ },
143
+ {
144
+ "epoch": 0.13,
145
+ "eval_loss": 1.761339783668518,
146
+ "eval_runtime": 37.661,
147
+ "eval_samples_per_second": 2.23,
148
+ "eval_steps_per_second": 0.292,
149
+ "step": 50
150
+ },
151
+ {
152
+ "epoch": 0.14,
153
+ "learning_rate": 9.29305912596401e-05,
154
+ "loss": 1.7305,
155
+ "step": 55
156
+ },
157
+ {
158
+ "epoch": 0.14,
159
+ "eval_loss": 1.7389689683914185,
160
+ "eval_runtime": 38.0055,
161
+ "eval_samples_per_second": 2.21,
162
+ "eval_steps_per_second": 0.289,
163
+ "step": 55
164
+ },
165
+ {
166
+ "epoch": 0.15,
167
+ "learning_rate": 9.228791773778921e-05,
168
+ "loss": 1.7086,
169
+ "step": 60
170
+ },
171
+ {
172
+ "epoch": 0.15,
173
+ "eval_loss": 1.713218092918396,
174
+ "eval_runtime": 37.737,
175
+ "eval_samples_per_second": 2.226,
176
+ "eval_steps_per_second": 0.291,
177
+ "step": 60
178
+ },
179
+ {
180
+ "epoch": 0.17,
181
+ "learning_rate": 9.16452442159383e-05,
182
+ "loss": 1.7057,
183
+ "step": 65
184
+ },
185
+ {
186
+ "epoch": 0.17,
187
+ "eval_loss": 1.7108122110366821,
188
+ "eval_runtime": 38.0037,
189
+ "eval_samples_per_second": 2.21,
190
+ "eval_steps_per_second": 0.289,
191
+ "step": 65
192
+ },
193
+ {
194
+ "epoch": 0.18,
195
+ "learning_rate": 9.100257069408741e-05,
196
+ "loss": 1.7243,
197
+ "step": 70
198
+ },
199
+ {
200
+ "epoch": 0.18,
201
+ "eval_loss": 1.7041622400283813,
202
+ "eval_runtime": 38.1402,
203
+ "eval_samples_per_second": 2.202,
204
+ "eval_steps_per_second": 0.288,
205
+ "step": 70
206
+ },
207
+ {
208
+ "epoch": 0.19,
209
+ "learning_rate": 9.03598971722365e-05,
210
+ "loss": 1.7038,
211
+ "step": 75
212
+ },
213
+ {
214
+ "epoch": 0.19,
215
+ "eval_loss": 1.7010408639907837,
216
+ "eval_runtime": 37.7193,
217
+ "eval_samples_per_second": 2.227,
218
+ "eval_steps_per_second": 0.292,
219
+ "step": 75
220
+ },
221
+ {
222
+ "epoch": 0.21,
223
+ "learning_rate": 8.97172236503856e-05,
224
+ "loss": 1.6939,
225
+ "step": 80
226
+ },
227
+ {
228
+ "epoch": 0.21,
229
+ "eval_loss": 1.6987501382827759,
230
+ "eval_runtime": 38.0624,
231
+ "eval_samples_per_second": 2.207,
232
+ "eval_steps_per_second": 0.289,
233
+ "step": 80
234
+ },
235
+ {
236
+ "epoch": 0.22,
237
+ "learning_rate": 8.907455012853471e-05,
238
+ "loss": 1.645,
239
+ "step": 85
240
+ },
241
+ {
242
+ "epoch": 0.22,
243
+ "eval_loss": 1.6968218088150024,
244
+ "eval_runtime": 37.9205,
245
+ "eval_samples_per_second": 2.215,
246
+ "eval_steps_per_second": 0.29,
247
+ "step": 85
248
+ },
249
+ {
250
+ "epoch": 0.23,
251
+ "learning_rate": 8.84318766066838e-05,
252
+ "loss": 1.7016,
253
+ "step": 90
254
+ },
255
+ {
256
+ "epoch": 0.23,
257
+ "eval_loss": 1.6952500343322754,
258
+ "eval_runtime": 37.9617,
259
+ "eval_samples_per_second": 2.213,
260
+ "eval_steps_per_second": 0.29,
261
+ "step": 90
262
+ },
263
+ {
264
+ "epoch": 0.24,
265
+ "learning_rate": 8.778920308483291e-05,
266
+ "loss": 1.6869,
267
+ "step": 95
268
+ },
269
+ {
270
+ "epoch": 0.24,
271
+ "eval_loss": 1.6932854652404785,
272
+ "eval_runtime": 37.8781,
273
+ "eval_samples_per_second": 2.218,
274
+ "eval_steps_per_second": 0.29,
275
+ "step": 95
276
+ },
277
+ {
278
+ "epoch": 0.26,
279
+ "learning_rate": 8.7146529562982e-05,
280
+ "loss": 1.6728,
281
+ "step": 100
282
+ },
283
+ {
284
+ "epoch": 0.26,
285
+ "eval_loss": 1.6920738220214844,
286
+ "eval_runtime": 38.1996,
287
+ "eval_samples_per_second": 2.199,
288
+ "eval_steps_per_second": 0.288,
289
+ "step": 100
290
+ },
291
+ {
292
+ "epoch": 0.27,
293
+ "learning_rate": 8.650385604113111e-05,
294
+ "loss": 1.6674,
295
+ "step": 105
296
+ },
297
+ {
298
+ "epoch": 0.27,
299
+ "eval_loss": 1.6904469728469849,
300
+ "eval_runtime": 38.1252,
301
+ "eval_samples_per_second": 2.203,
302
+ "eval_steps_per_second": 0.289,
303
+ "step": 105
304
+ },
305
+ {
306
+ "epoch": 0.28,
307
+ "learning_rate": 8.586118251928022e-05,
308
+ "loss": 1.7013,
309
+ "step": 110
310
+ },
311
+ {
312
+ "epoch": 0.28,
313
+ "eval_loss": 1.6895124912261963,
314
+ "eval_runtime": 38.0968,
315
+ "eval_samples_per_second": 2.205,
316
+ "eval_steps_per_second": 0.289,
317
+ "step": 110
318
+ },
319
+ {
320
+ "epoch": 0.3,
321
+ "learning_rate": 8.521850899742931e-05,
322
+ "loss": 1.7148,
323
+ "step": 115
324
+ },
325
+ {
326
+ "epoch": 0.3,
327
+ "eval_loss": 1.6886374950408936,
328
+ "eval_runtime": 37.8799,
329
+ "eval_samples_per_second": 2.218,
330
+ "eval_steps_per_second": 0.29,
331
+ "step": 115
332
+ },
333
+ {
334
+ "epoch": 0.31,
335
+ "learning_rate": 8.457583547557842e-05,
336
+ "loss": 1.7166,
337
+ "step": 120
338
+ },
339
+ {
340
+ "epoch": 0.31,
341
+ "eval_loss": 1.6868164539337158,
342
+ "eval_runtime": 37.8861,
343
+ "eval_samples_per_second": 2.217,
344
+ "eval_steps_per_second": 0.29,
345
+ "step": 120
346
+ },
347
+ {
348
+ "epoch": 0.32,
349
+ "learning_rate": 8.393316195372751e-05,
350
+ "loss": 1.7012,
351
+ "step": 125
352
+ },
353
+ {
354
+ "epoch": 0.32,
355
+ "eval_loss": 1.6858367919921875,
356
+ "eval_runtime": 38.1136,
357
+ "eval_samples_per_second": 2.204,
358
+ "eval_steps_per_second": 0.289,
359
+ "step": 125
360
+ },
361
+ {
362
+ "epoch": 0.33,
363
+ "learning_rate": 8.32904884318766e-05,
364
+ "loss": 1.6827,
365
+ "step": 130
366
+ },
367
+ {
368
+ "epoch": 0.33,
369
+ "eval_loss": 1.6849361658096313,
370
+ "eval_runtime": 37.9883,
371
+ "eval_samples_per_second": 2.211,
372
+ "eval_steps_per_second": 0.29,
373
+ "step": 130
374
+ },
375
+ {
376
+ "epoch": 0.35,
377
+ "learning_rate": 8.264781491002571e-05,
378
+ "loss": 1.7157,
379
+ "step": 135
380
+ },
381
+ {
382
+ "epoch": 0.35,
383
+ "eval_loss": 1.684756875038147,
384
+ "eval_runtime": 37.8726,
385
+ "eval_samples_per_second": 2.218,
386
+ "eval_steps_per_second": 0.29,
387
+ "step": 135
388
+ },
389
+ {
390
+ "epoch": 0.36,
391
+ "learning_rate": 8.200514138817481e-05,
392
+ "loss": 1.6668,
393
+ "step": 140
394
+ },
395
+ {
396
+ "epoch": 0.36,
397
+ "eval_loss": 1.6836236715316772,
398
+ "eval_runtime": 37.9586,
399
+ "eval_samples_per_second": 2.213,
400
+ "eval_steps_per_second": 0.29,
401
+ "step": 140
402
+ },
403
+ {
404
+ "epoch": 0.37,
405
+ "learning_rate": 8.136246786632391e-05,
406
+ "loss": 1.6588,
407
+ "step": 145
408
+ },
409
+ {
410
+ "epoch": 0.37,
411
+ "eval_loss": 1.6828166246414185,
412
+ "eval_runtime": 38.0596,
413
+ "eval_samples_per_second": 2.207,
414
+ "eval_steps_per_second": 0.289,
415
+ "step": 145
416
+ },
417
+ {
418
+ "epoch": 0.39,
419
+ "learning_rate": 8.071979434447301e-05,
420
+ "loss": 1.7005,
421
+ "step": 150
422
+ },
423
+ {
424
+ "epoch": 0.39,
425
+ "eval_loss": 1.681463360786438,
426
+ "eval_runtime": 38.0878,
427
+ "eval_samples_per_second": 2.205,
428
+ "eval_steps_per_second": 0.289,
429
+ "step": 150
430
+ },
431
+ {
432
+ "epoch": 0.4,
433
+ "learning_rate": 8.007712082262212e-05,
434
+ "loss": 1.6893,
435
+ "step": 155
436
+ },
437
+ {
438
+ "epoch": 0.4,
439
+ "eval_loss": 1.6811552047729492,
440
+ "eval_runtime": 37.9265,
441
+ "eval_samples_per_second": 2.215,
442
+ "eval_steps_per_second": 0.29,
443
+ "step": 155
444
+ },
445
+ {
446
+ "epoch": 0.41,
447
+ "learning_rate": 7.943444730077121e-05,
448
+ "loss": 1.6682,
449
+ "step": 160
450
+ },
451
+ {
452
+ "epoch": 0.41,
453
+ "eval_loss": 1.6801096200942993,
454
+ "eval_runtime": 38.0375,
455
+ "eval_samples_per_second": 2.208,
456
+ "eval_steps_per_second": 0.289,
457
+ "step": 160
458
+ },
459
+ {
460
+ "epoch": 0.42,
461
+ "learning_rate": 7.87917737789203e-05,
462
+ "loss": 1.6481,
463
+ "step": 165
464
+ },
465
+ {
466
+ "epoch": 0.42,
467
+ "eval_loss": 1.6800144910812378,
468
+ "eval_runtime": 37.8916,
469
+ "eval_samples_per_second": 2.217,
470
+ "eval_steps_per_second": 0.29,
471
+ "step": 165
472
+ },
473
+ {
474
+ "epoch": 0.44,
475
+ "learning_rate": 7.814910025706941e-05,
476
+ "loss": 1.7042,
477
+ "step": 170
478
+ },
479
+ {
480
+ "epoch": 0.44,
481
+ "eval_loss": 1.6784976720809937,
482
+ "eval_runtime": 37.8514,
483
+ "eval_samples_per_second": 2.219,
484
+ "eval_steps_per_second": 0.291,
485
+ "step": 170
486
+ },
487
+ {
488
+ "epoch": 0.45,
489
+ "learning_rate": 7.750642673521852e-05,
490
+ "loss": 1.6555,
491
+ "step": 175
492
+ },
493
+ {
494
+ "epoch": 0.45,
495
+ "eval_loss": 1.6780468225479126,
496
+ "eval_runtime": 37.8893,
497
+ "eval_samples_per_second": 2.217,
498
+ "eval_steps_per_second": 0.29,
499
+ "step": 175
500
+ },
501
+ {
502
+ "epoch": 0.46,
503
+ "learning_rate": 7.686375321336761e-05,
504
+ "loss": 1.6717,
505
+ "step": 180
506
+ },
507
+ {
508
+ "epoch": 0.46,
509
+ "eval_loss": 1.6775085926055908,
510
+ "eval_runtime": 37.9289,
511
+ "eval_samples_per_second": 2.215,
512
+ "eval_steps_per_second": 0.29,
513
+ "step": 180
514
+ },
515
+ {
516
+ "epoch": 0.48,
517
+ "learning_rate": 7.622107969151672e-05,
518
+ "loss": 1.6716,
519
+ "step": 185
520
+ },
521
+ {
522
+ "epoch": 0.48,
523
+ "eval_loss": 1.6770914793014526,
524
+ "eval_runtime": 38.0111,
525
+ "eval_samples_per_second": 2.21,
526
+ "eval_steps_per_second": 0.289,
527
+ "step": 185
528
+ },
529
+ {
530
+ "epoch": 0.49,
531
+ "learning_rate": 7.557840616966581e-05,
532
+ "loss": 1.6618,
533
+ "step": 190
534
+ },
535
+ {
536
+ "epoch": 0.49,
537
+ "eval_loss": 1.676563024520874,
538
+ "eval_runtime": 38.0703,
539
+ "eval_samples_per_second": 2.206,
540
+ "eval_steps_per_second": 0.289,
541
+ "step": 190
542
+ },
543
+ {
544
+ "epoch": 0.5,
545
+ "learning_rate": 7.493573264781492e-05,
546
+ "loss": 1.6411,
547
+ "step": 195
548
+ },
549
+ {
550
+ "epoch": 0.5,
551
+ "eval_loss": 1.6757832765579224,
552
+ "eval_runtime": 37.827,
553
+ "eval_samples_per_second": 2.221,
554
+ "eval_steps_per_second": 0.291,
555
+ "step": 195
556
+ },
557
+ {
558
+ "epoch": 0.51,
559
+ "learning_rate": 7.429305912596401e-05,
560
+ "loss": 1.6637,
561
+ "step": 200
562
+ },
563
+ {
564
+ "epoch": 0.51,
565
+ "eval_loss": 1.6751985549926758,
566
+ "eval_runtime": 37.836,
567
+ "eval_samples_per_second": 2.22,
568
+ "eval_steps_per_second": 0.291,
569
+ "step": 200
570
+ },
571
+ {
572
+ "epoch": 0.53,
573
+ "learning_rate": 7.365038560411311e-05,
574
+ "loss": 1.6142,
575
+ "step": 205
576
+ },
577
+ {
578
+ "epoch": 0.53,
579
+ "eval_loss": 1.675147294998169,
580
+ "eval_runtime": 37.9175,
581
+ "eval_samples_per_second": 2.215,
582
+ "eval_steps_per_second": 0.29,
583
+ "step": 205
584
+ },
585
+ {
586
+ "epoch": 0.54,
587
+ "learning_rate": 7.300771208226222e-05,
588
+ "loss": 1.6741,
589
+ "step": 210
590
+ },
591
+ {
592
+ "epoch": 0.54,
593
+ "eval_loss": 1.674437165260315,
594
+ "eval_runtime": 38.087,
595
+ "eval_samples_per_second": 2.205,
596
+ "eval_steps_per_second": 0.289,
597
+ "step": 210
598
+ },
599
+ {
600
+ "epoch": 0.55,
601
+ "learning_rate": 7.236503856041131e-05,
602
+ "loss": 1.6408,
603
+ "step": 215
604
+ },
605
+ {
606
+ "epoch": 0.55,
607
+ "eval_loss": 1.6737719774246216,
608
+ "eval_runtime": 38.0075,
609
+ "eval_samples_per_second": 2.21,
610
+ "eval_steps_per_second": 0.289,
611
+ "step": 215
612
+ },
613
+ {
614
+ "epoch": 0.57,
615
+ "learning_rate": 7.172236503856042e-05,
616
+ "loss": 1.6733,
617
+ "step": 220
618
+ },
619
+ {
620
+ "epoch": 0.57,
621
+ "eval_loss": 1.6732759475708008,
622
+ "eval_runtime": 38.1125,
623
+ "eval_samples_per_second": 2.204,
624
+ "eval_steps_per_second": 0.289,
625
+ "step": 220
626
+ },
627
+ {
628
+ "epoch": 0.58,
629
+ "learning_rate": 7.107969151670951e-05,
630
+ "loss": 1.679,
631
+ "step": 225
632
+ },
633
+ {
634
+ "epoch": 0.58,
635
+ "eval_loss": 1.6726195812225342,
636
+ "eval_runtime": 37.9769,
637
+ "eval_samples_per_second": 2.212,
638
+ "eval_steps_per_second": 0.29,
639
+ "step": 225
640
+ },
641
+ {
642
+ "epoch": 0.59,
643
+ "learning_rate": 7.043701799485862e-05,
644
+ "loss": 1.7202,
645
+ "step": 230
646
+ },
647
+ {
648
+ "epoch": 0.59,
649
+ "eval_loss": 1.671908974647522,
650
+ "eval_runtime": 37.8215,
651
+ "eval_samples_per_second": 2.221,
652
+ "eval_steps_per_second": 0.291,
653
+ "step": 230
654
+ },
655
+ {
656
+ "epoch": 0.6,
657
+ "learning_rate": 6.979434447300771e-05,
658
+ "loss": 1.6805,
659
+ "step": 235
660
+ },
661
+ {
662
+ "epoch": 0.6,
663
+ "eval_loss": 1.6715577840805054,
664
+ "eval_runtime": 37.9343,
665
+ "eval_samples_per_second": 2.214,
666
+ "eval_steps_per_second": 0.29,
667
+ "step": 235
668
+ },
669
+ {
670
+ "epoch": 0.62,
671
+ "learning_rate": 6.91516709511568e-05,
672
+ "loss": 1.6331,
673
+ "step": 240
674
+ },
675
+ {
676
+ "epoch": 0.62,
677
+ "eval_loss": 1.6715376377105713,
678
+ "eval_runtime": 37.9904,
679
+ "eval_samples_per_second": 2.211,
680
+ "eval_steps_per_second": 0.29,
681
+ "step": 240
682
+ },
683
+ {
684
+ "epoch": 0.63,
685
+ "learning_rate": 6.850899742930593e-05,
686
+ "loss": 1.6761,
687
+ "step": 245
688
+ },
689
+ {
690
+ "epoch": 0.63,
691
+ "eval_loss": 1.671446681022644,
692
+ "eval_runtime": 38.0357,
693
+ "eval_samples_per_second": 2.208,
694
+ "eval_steps_per_second": 0.289,
695
+ "step": 245
696
+ },
697
+ {
698
+ "epoch": 0.64,
699
+ "learning_rate": 6.786632390745502e-05,
700
+ "loss": 1.6994,
701
+ "step": 250
702
+ },
703
+ {
704
+ "epoch": 0.64,
705
+ "eval_loss": 1.6716604232788086,
706
+ "eval_runtime": 37.9609,
707
+ "eval_samples_per_second": 2.213,
708
+ "eval_steps_per_second": 0.29,
709
+ "step": 250
710
+ },
711
+ {
712
+ "epoch": 0.66,
713
+ "learning_rate": 6.722365038560411e-05,
714
+ "loss": 1.6305,
715
+ "step": 255
716
+ },
717
+ {
718
+ "epoch": 0.66,
719
+ "eval_loss": 1.6711723804473877,
720
+ "eval_runtime": 37.7893,
721
+ "eval_samples_per_second": 2.223,
722
+ "eval_steps_per_second": 0.291,
723
+ "step": 255
724
+ },
725
+ {
726
+ "epoch": 0.67,
727
+ "learning_rate": 6.658097686375322e-05,
728
+ "loss": 1.6612,
729
+ "step": 260
730
+ },
731
+ {
732
+ "epoch": 0.67,
733
+ "eval_loss": 1.670398473739624,
734
+ "eval_runtime": 37.7667,
735
+ "eval_samples_per_second": 2.224,
736
+ "eval_steps_per_second": 0.291,
737
+ "step": 260
738
+ },
739
+ {
740
+ "epoch": 0.68,
741
+ "learning_rate": 6.593830334190231e-05,
742
+ "loss": 1.6576,
743
+ "step": 265
744
+ },
745
+ {
746
+ "epoch": 0.68,
747
+ "eval_loss": 1.6706459522247314,
748
+ "eval_runtime": 37.8285,
749
+ "eval_samples_per_second": 2.221,
750
+ "eval_steps_per_second": 0.291,
751
+ "step": 265
752
+ },
753
+ {
754
+ "epoch": 0.69,
755
+ "learning_rate": 6.529562982005142e-05,
756
+ "loss": 1.6837,
757
+ "step": 270
758
+ },
759
+ {
760
+ "epoch": 0.69,
761
+ "eval_loss": 1.6699285507202148,
762
+ "eval_runtime": 37.7619,
763
+ "eval_samples_per_second": 2.224,
764
+ "eval_steps_per_second": 0.291,
765
+ "step": 270
766
+ },
767
+ {
768
+ "epoch": 0.71,
769
+ "learning_rate": 6.465295629820052e-05,
770
+ "loss": 1.6493,
771
+ "step": 275
772
+ },
773
+ {
774
+ "epoch": 0.71,
775
+ "eval_loss": 1.6693135499954224,
776
+ "eval_runtime": 37.7441,
777
+ "eval_samples_per_second": 2.226,
778
+ "eval_steps_per_second": 0.291,
779
+ "step": 275
780
+ },
781
+ {
782
+ "epoch": 0.72,
783
+ "learning_rate": 6.401028277634962e-05,
784
+ "loss": 1.6427,
785
+ "step": 280
786
+ },
787
+ {
788
+ "epoch": 0.72,
789
+ "eval_loss": 1.669467568397522,
790
+ "eval_runtime": 37.8707,
791
+ "eval_samples_per_second": 2.218,
792
+ "eval_steps_per_second": 0.29,
793
+ "step": 280
794
+ },
795
+ {
796
+ "epoch": 0.73,
797
+ "learning_rate": 6.336760925449872e-05,
798
+ "loss": 1.6236,
799
+ "step": 285
800
+ },
801
+ {
802
+ "epoch": 0.73,
803
+ "eval_loss": 1.6690431833267212,
804
+ "eval_runtime": 37.8946,
805
+ "eval_samples_per_second": 2.217,
806
+ "eval_steps_per_second": 0.29,
807
+ "step": 285
808
+ },
809
+ {
810
+ "epoch": 0.75,
811
+ "learning_rate": 6.272493573264781e-05,
812
+ "loss": 1.6538,
813
+ "step": 290
814
+ },
815
+ {
816
+ "epoch": 0.75,
817
+ "eval_loss": 1.6688117980957031,
818
+ "eval_runtime": 37.9378,
819
+ "eval_samples_per_second": 2.214,
820
+ "eval_steps_per_second": 0.29,
821
+ "step": 290
822
+ },
823
+ {
824
+ "epoch": 0.76,
825
+ "learning_rate": 6.208226221079692e-05,
826
+ "loss": 1.6558,
827
+ "step": 295
828
+ },
829
+ {
830
+ "epoch": 0.76,
831
+ "eval_loss": 1.668560266494751,
832
+ "eval_runtime": 38.0448,
833
+ "eval_samples_per_second": 2.208,
834
+ "eval_steps_per_second": 0.289,
835
+ "step": 295
836
+ },
837
+ {
838
+ "epoch": 0.77,
839
+ "learning_rate": 6.143958868894601e-05,
840
+ "loss": 1.6401,
841
+ "step": 300
842
+ },
843
+ {
844
+ "epoch": 0.77,
845
+ "eval_loss": 1.6680612564086914,
846
+ "eval_runtime": 37.8571,
847
+ "eval_samples_per_second": 2.219,
848
+ "eval_steps_per_second": 0.291,
849
+ "step": 300
850
+ },
851
+ {
852
+ "epoch": 0.78,
853
+ "learning_rate": 6.079691516709511e-05,
854
+ "loss": 1.6468,
855
+ "step": 305
856
+ },
857
+ {
858
+ "epoch": 0.78,
859
+ "eval_loss": 1.6675914525985718,
860
+ "eval_runtime": 37.8491,
861
+ "eval_samples_per_second": 2.219,
862
+ "eval_steps_per_second": 0.291,
863
+ "step": 305
864
+ },
865
+ {
866
+ "epoch": 0.8,
867
+ "learning_rate": 6.015424164524421e-05,
868
+ "loss": 1.6579,
869
+ "step": 310
870
+ },
871
+ {
872
+ "epoch": 0.8,
873
+ "eval_loss": 1.6671632528305054,
874
+ "eval_runtime": 37.7658,
875
+ "eval_samples_per_second": 2.224,
876
+ "eval_steps_per_second": 0.291,
877
+ "step": 310
878
+ },
879
+ {
880
+ "epoch": 0.81,
881
+ "learning_rate": 5.951156812339333e-05,
882
+ "loss": 1.6339,
883
+ "step": 315
884
+ },
885
+ {
886
+ "epoch": 0.81,
887
+ "eval_loss": 1.666803240776062,
888
+ "eval_runtime": 37.7896,
889
+ "eval_samples_per_second": 2.223,
890
+ "eval_steps_per_second": 0.291,
891
+ "step": 315
892
+ },
893
+ {
894
+ "epoch": 0.82,
895
+ "learning_rate": 5.886889460154242e-05,
896
+ "loss": 1.6636,
897
+ "step": 320
898
+ },
899
+ {
900
+ "epoch": 0.82,
901
+ "eval_loss": 1.6664105653762817,
902
+ "eval_runtime": 38.0586,
903
+ "eval_samples_per_second": 2.207,
904
+ "eval_steps_per_second": 0.289,
905
+ "step": 320
906
+ },
907
+ {
908
+ "epoch": 0.84,
909
+ "learning_rate": 5.822622107969152e-05,
910
+ "loss": 1.641,
911
+ "step": 325
912
+ },
913
+ {
914
+ "epoch": 0.84,
915
+ "eval_loss": 1.6665875911712646,
916
+ "eval_runtime": 37.9046,
917
+ "eval_samples_per_second": 2.216,
918
+ "eval_steps_per_second": 0.29,
919
+ "step": 325
920
+ },
921
+ {
922
+ "epoch": 0.85,
923
+ "learning_rate": 5.758354755784062e-05,
924
+ "loss": 1.6616,
925
+ "step": 330
926
+ },
927
+ {
928
+ "epoch": 0.85,
929
+ "eval_loss": 1.6664165258407593,
930
+ "eval_runtime": 37.8575,
931
+ "eval_samples_per_second": 2.219,
932
+ "eval_steps_per_second": 0.291,
933
+ "step": 330
934
+ },
935
+ {
936
+ "epoch": 0.86,
937
+ "learning_rate": 5.694087403598972e-05,
938
+ "loss": 1.6597,
939
+ "step": 335
940
+ },
941
+ {
942
+ "epoch": 0.86,
943
+ "eval_loss": 1.666045904159546,
944
+ "eval_runtime": 37.9711,
945
+ "eval_samples_per_second": 2.212,
946
+ "eval_steps_per_second": 0.29,
947
+ "step": 335
948
+ },
949
+ {
950
+ "epoch": 0.87,
951
+ "learning_rate": 5.6298200514138824e-05,
952
+ "loss": 1.67,
953
+ "step": 340
954
+ },
955
+ {
956
+ "epoch": 0.87,
957
+ "eval_loss": 1.666337490081787,
958
+ "eval_runtime": 37.8821,
959
+ "eval_samples_per_second": 2.217,
960
+ "eval_steps_per_second": 0.29,
961
+ "step": 340
962
+ },
963
+ {
964
+ "epoch": 0.89,
965
+ "learning_rate": 5.5655526992287924e-05,
966
+ "loss": 1.6344,
967
+ "step": 345
968
+ },
969
+ {
970
+ "epoch": 0.89,
971
+ "eval_loss": 1.6659029722213745,
972
+ "eval_runtime": 37.9188,
973
+ "eval_samples_per_second": 2.215,
974
+ "eval_steps_per_second": 0.29,
975
+ "step": 345
976
+ },
977
+ {
978
+ "epoch": 0.9,
979
+ "learning_rate": 5.501285347043702e-05,
980
+ "loss": 1.6623,
981
+ "step": 350
982
+ },
983
+ {
984
+ "epoch": 0.9,
985
+ "eval_loss": 1.6656525135040283,
986
+ "eval_runtime": 37.9951,
987
+ "eval_samples_per_second": 2.211,
988
+ "eval_steps_per_second": 0.29,
989
+ "step": 350
990
+ },
991
+ {
992
+ "epoch": 0.91,
993
+ "learning_rate": 5.437017994858612e-05,
994
+ "loss": 1.6623,
995
+ "step": 355
996
+ },
997
+ {
998
+ "epoch": 0.91,
999
+ "eval_loss": 1.6654421091079712,
1000
+ "eval_runtime": 37.8803,
1001
+ "eval_samples_per_second": 2.218,
1002
+ "eval_steps_per_second": 0.29,
1003
+ "step": 355
1004
+ },
1005
+ {
1006
+ "epoch": 0.93,
1007
+ "learning_rate": 5.372750642673522e-05,
1008
+ "loss": 1.6741,
1009
+ "step": 360
1010
+ },
1011
+ {
1012
+ "epoch": 0.93,
1013
+ "eval_loss": 1.665541648864746,
1014
+ "eval_runtime": 37.7151,
1015
+ "eval_samples_per_second": 2.227,
1016
+ "eval_steps_per_second": 0.292,
1017
+ "step": 360
1018
+ },
1019
+ {
1020
+ "epoch": 0.94,
1021
+ "learning_rate": 5.308483290488432e-05,
1022
+ "loss": 1.6973,
1023
+ "step": 365
1024
+ },
1025
+ {
1026
+ "epoch": 0.94,
1027
+ "eval_loss": 1.6651471853256226,
1028
+ "eval_runtime": 37.9144,
1029
+ "eval_samples_per_second": 2.216,
1030
+ "eval_steps_per_second": 0.29,
1031
+ "step": 365
1032
+ },
1033
+ {
1034
+ "epoch": 0.95,
1035
+ "learning_rate": 5.244215938303342e-05,
1036
+ "loss": 1.6829,
1037
+ "step": 370
1038
+ },
1039
+ {
1040
+ "epoch": 0.95,
1041
+ "eval_loss": 1.6651064157485962,
1042
+ "eval_runtime": 37.7443,
1043
+ "eval_samples_per_second": 2.226,
1044
+ "eval_steps_per_second": 0.291,
1045
+ "step": 370
1046
+ },
1047
+ {
1048
+ "epoch": 0.96,
1049
+ "learning_rate": 5.1799485861182514e-05,
1050
+ "loss": 1.6923,
1051
+ "step": 375
1052
+ },
1053
+ {
1054
+ "epoch": 0.96,
1055
+ "eval_loss": 1.6646034717559814,
1056
+ "eval_runtime": 37.8783,
1057
+ "eval_samples_per_second": 2.218,
1058
+ "eval_steps_per_second": 0.29,
1059
+ "step": 375
1060
+ },
1061
+ {
1062
+ "epoch": 0.98,
1063
+ "learning_rate": 5.1156812339331615e-05,
1064
+ "loss": 1.6725,
1065
+ "step": 380
1066
+ },
1067
+ {
1068
+ "epoch": 0.98,
1069
+ "eval_loss": 1.6639235019683838,
1070
+ "eval_runtime": 37.7559,
1071
+ "eval_samples_per_second": 2.225,
1072
+ "eval_steps_per_second": 0.291,
1073
+ "step": 380
1074
+ },
1075
+ {
1076
+ "epoch": 0.99,
1077
+ "learning_rate": 5.051413881748073e-05,
1078
+ "loss": 1.6216,
1079
+ "step": 385
1080
+ },
1081
+ {
1082
+ "epoch": 0.99,
1083
+ "eval_loss": 1.6643506288528442,
1084
+ "eval_runtime": 37.9895,
1085
+ "eval_samples_per_second": 2.211,
1086
+ "eval_steps_per_second": 0.29,
1087
+ "step": 385
1088
+ },
1089
+ {
1090
+ "epoch": 1.0,
1091
+ "learning_rate": 4.987146529562982e-05,
1092
+ "loss": 1.6518,
1093
+ "step": 390
1094
+ },
1095
+ {
1096
+ "epoch": 1.0,
1097
+ "eval_loss": 1.6643046140670776,
1098
+ "eval_runtime": 37.9289,
1099
+ "eval_samples_per_second": 2.215,
1100
+ "eval_steps_per_second": 0.29,
1101
+ "step": 390
1102
+ },
1103
+ {
1104
+ "epoch": 1.02,
1105
+ "learning_rate": 4.922879177377892e-05,
1106
+ "loss": 1.658,
1107
+ "step": 395
1108
+ },
1109
+ {
1110
+ "epoch": 1.02,
1111
+ "eval_loss": 1.6641957759857178,
1112
+ "eval_runtime": 37.8638,
1113
+ "eval_samples_per_second": 2.218,
1114
+ "eval_steps_per_second": 0.291,
1115
+ "step": 395
1116
+ },
1117
+ {
1118
+ "epoch": 1.03,
1119
+ "learning_rate": 4.8586118251928024e-05,
1120
+ "loss": 1.6767,
1121
+ "step": 400
1122
+ },
1123
+ {
1124
+ "epoch": 1.03,
1125
+ "eval_loss": 1.663757085800171,
1126
+ "eval_runtime": 37.7728,
1127
+ "eval_samples_per_second": 2.224,
1128
+ "eval_steps_per_second": 0.291,
1129
+ "step": 400
1130
+ },
1131
+ {
1132
+ "epoch": 1.04,
1133
+ "learning_rate": 4.7943444730077124e-05,
1134
+ "loss": 1.6264,
1135
+ "step": 405
1136
+ },
1137
+ {
1138
+ "epoch": 1.04,
1139
+ "eval_loss": 1.6635041236877441,
1140
+ "eval_runtime": 37.7702,
1141
+ "eval_samples_per_second": 2.224,
1142
+ "eval_steps_per_second": 0.291,
1143
+ "step": 405
1144
+ },
1145
+ {
1146
+ "epoch": 1.05,
1147
+ "learning_rate": 4.7300771208226225e-05,
1148
+ "loss": 1.6527,
1149
+ "step": 410
1150
+ },
1151
+ {
1152
+ "epoch": 1.05,
1153
+ "eval_loss": 1.6632280349731445,
1154
+ "eval_runtime": 37.8453,
1155
+ "eval_samples_per_second": 2.22,
1156
+ "eval_steps_per_second": 0.291,
1157
+ "step": 410
1158
+ },
1159
+ {
1160
+ "epoch": 1.07,
1161
+ "learning_rate": 4.6658097686375325e-05,
1162
+ "loss": 1.6157,
1163
+ "step": 415
1164
+ },
1165
+ {
1166
+ "epoch": 1.07,
1167
+ "eval_loss": 1.6636598110198975,
1168
+ "eval_runtime": 37.812,
1169
+ "eval_samples_per_second": 2.222,
1170
+ "eval_steps_per_second": 0.291,
1171
+ "step": 415
1172
+ },
1173
+ {
1174
+ "epoch": 1.08,
1175
+ "learning_rate": 4.6015424164524426e-05,
1176
+ "loss": 1.5966,
1177
+ "step": 420
1178
+ },
1179
+ {
1180
+ "epoch": 1.08,
1181
+ "eval_loss": 1.663393497467041,
1182
+ "eval_runtime": 37.7743,
1183
+ "eval_samples_per_second": 2.224,
1184
+ "eval_steps_per_second": 0.291,
1185
+ "step": 420
1186
+ },
1187
+ {
1188
+ "epoch": 1.09,
1189
+ "learning_rate": 4.537275064267352e-05,
1190
+ "loss": 1.6705,
1191
+ "step": 425
1192
+ },
1193
+ {
1194
+ "epoch": 1.09,
1195
+ "eval_loss": 1.6631269454956055,
1196
+ "eval_runtime": 38.0953,
1197
+ "eval_samples_per_second": 2.205,
1198
+ "eval_steps_per_second": 0.289,
1199
+ "step": 425
1200
+ },
1201
+ {
1202
+ "epoch": 1.11,
1203
+ "learning_rate": 4.473007712082262e-05,
1204
+ "loss": 1.6691,
1205
+ "step": 430
1206
+ },
1207
+ {
1208
+ "epoch": 1.11,
1209
+ "eval_loss": 1.6633110046386719,
1210
+ "eval_runtime": 37.8964,
1211
+ "eval_samples_per_second": 2.217,
1212
+ "eval_steps_per_second": 0.29,
1213
+ "step": 430
1214
+ },
1215
+ {
1216
+ "epoch": 1.12,
1217
+ "learning_rate": 4.408740359897173e-05,
1218
+ "loss": 1.6332,
1219
+ "step": 435
1220
+ },
1221
+ {
1222
+ "epoch": 1.12,
1223
+ "eval_loss": 1.6628649234771729,
1224
+ "eval_runtime": 38.0621,
1225
+ "eval_samples_per_second": 2.207,
1226
+ "eval_steps_per_second": 0.289,
1227
+ "step": 435
1228
+ },
1229
+ {
1230
+ "epoch": 1.13,
1231
+ "learning_rate": 4.344473007712083e-05,
1232
+ "loss": 1.5916,
1233
+ "step": 440
1234
+ },
1235
+ {
1236
+ "epoch": 1.13,
1237
+ "eval_loss": 1.662834882736206,
1238
+ "eval_runtime": 37.9722,
1239
+ "eval_samples_per_second": 2.212,
1240
+ "eval_steps_per_second": 0.29,
1241
+ "step": 440
1242
+ },
1243
+ {
1244
+ "epoch": 1.14,
1245
+ "learning_rate": 4.280205655526993e-05,
1246
+ "loss": 1.6543,
1247
+ "step": 445
1248
+ },
1249
+ {
1250
+ "epoch": 1.14,
1251
+ "eval_loss": 1.6630265712738037,
1252
+ "eval_runtime": 37.9699,
1253
+ "eval_samples_per_second": 2.212,
1254
+ "eval_steps_per_second": 0.29,
1255
+ "step": 445
1256
+ },
1257
+ {
1258
+ "epoch": 1.16,
1259
+ "learning_rate": 4.215938303341902e-05,
1260
+ "loss": 1.6353,
1261
+ "step": 450
1262
+ },
1263
+ {
1264
+ "epoch": 1.16,
1265
+ "eval_loss": 1.6625033617019653,
1266
+ "eval_runtime": 37.7505,
1267
+ "eval_samples_per_second": 2.225,
1268
+ "eval_steps_per_second": 0.291,
1269
+ "step": 450
1270
+ },
1271
+ {
1272
+ "epoch": 1.17,
1273
+ "learning_rate": 4.151670951156812e-05,
1274
+ "loss": 1.6441,
1275
+ "step": 455
1276
+ },
1277
+ {
1278
+ "epoch": 1.17,
1279
+ "eval_loss": 1.662785530090332,
1280
+ "eval_runtime": 37.75,
1281
+ "eval_samples_per_second": 2.225,
1282
+ "eval_steps_per_second": 0.291,
1283
+ "step": 455
1284
+ },
1285
+ {
1286
+ "epoch": 1.18,
1287
+ "learning_rate": 4.0874035989717224e-05,
1288
+ "loss": 1.6631,
1289
+ "step": 460
1290
+ },
1291
+ {
1292
+ "epoch": 1.18,
1293
+ "eval_loss": 1.6627779006958008,
1294
+ "eval_runtime": 37.864,
1295
+ "eval_samples_per_second": 2.218,
1296
+ "eval_steps_per_second": 0.291,
1297
+ "step": 460
1298
+ },
1299
+ {
1300
+ "epoch": 1.2,
1301
+ "learning_rate": 4.0231362467866324e-05,
1302
+ "loss": 1.6327,
1303
+ "step": 465
1304
+ },
1305
+ {
1306
+ "epoch": 1.2,
1307
+ "eval_loss": 1.662458896636963,
1308
+ "eval_runtime": 37.7496,
1309
+ "eval_samples_per_second": 2.225,
1310
+ "eval_steps_per_second": 0.291,
1311
+ "step": 465
1312
+ },
1313
+ {
1314
+ "epoch": 1.21,
1315
+ "learning_rate": 3.958868894601543e-05,
1316
+ "loss": 1.6238,
1317
+ "step": 470
1318
+ },
1319
+ {
1320
+ "epoch": 1.21,
1321
+ "eval_loss": 1.6621876955032349,
1322
+ "eval_runtime": 37.8943,
1323
+ "eval_samples_per_second": 2.217,
1324
+ "eval_steps_per_second": 0.29,
1325
+ "step": 470
1326
+ },
1327
+ {
1328
+ "epoch": 1.22,
1329
+ "learning_rate": 3.8946015424164526e-05,
1330
+ "loss": 1.6231,
1331
+ "step": 475
1332
+ },
1333
+ {
1334
+ "epoch": 1.22,
1335
+ "eval_loss": 1.662013292312622,
1336
+ "eval_runtime": 37.9364,
1337
+ "eval_samples_per_second": 2.214,
1338
+ "eval_steps_per_second": 0.29,
1339
+ "step": 475
1340
+ },
1341
+ {
1342
+ "epoch": 1.23,
1343
+ "learning_rate": 3.8303341902313626e-05,
1344
+ "loss": 1.6381,
1345
+ "step": 480
1346
+ },
1347
+ {
1348
+ "epoch": 1.23,
1349
+ "eval_loss": 1.6615785360336304,
1350
+ "eval_runtime": 37.91,
1351
+ "eval_samples_per_second": 2.216,
1352
+ "eval_steps_per_second": 0.29,
1353
+ "step": 480
1354
+ },
1355
+ {
1356
+ "epoch": 1.25,
1357
+ "learning_rate": 3.766066838046273e-05,
1358
+ "loss": 1.6419,
1359
+ "step": 485
1360
+ },
1361
+ {
1362
+ "epoch": 1.25,
1363
+ "eval_loss": 1.662042498588562,
1364
+ "eval_runtime": 37.7834,
1365
+ "eval_samples_per_second": 2.223,
1366
+ "eval_steps_per_second": 0.291,
1367
+ "step": 485
1368
+ },
1369
+ {
1370
+ "epoch": 1.26,
1371
+ "learning_rate": 3.701799485861183e-05,
1372
+ "loss": 1.6319,
1373
+ "step": 490
1374
+ },
1375
+ {
1376
+ "epoch": 1.26,
1377
+ "eval_loss": 1.6619195938110352,
1378
+ "eval_runtime": 37.7385,
1379
+ "eval_samples_per_second": 2.226,
1380
+ "eval_steps_per_second": 0.291,
1381
+ "step": 490
1382
+ },
1383
+ {
1384
+ "epoch": 1.27,
1385
+ "learning_rate": 3.637532133676093e-05,
1386
+ "loss": 1.609,
1387
+ "step": 495
1388
+ },
1389
+ {
1390
+ "epoch": 1.27,
1391
+ "eval_loss": 1.6615225076675415,
1392
+ "eval_runtime": 37.8056,
1393
+ "eval_samples_per_second": 2.222,
1394
+ "eval_steps_per_second": 0.291,
1395
+ "step": 495
1396
+ },
1397
+ {
1398
+ "epoch": 1.29,
1399
+ "learning_rate": 3.573264781491003e-05,
1400
+ "loss": 1.6435,
1401
+ "step": 500
1402
+ },
1403
+ {
1404
+ "epoch": 1.29,
1405
+ "eval_loss": 1.6610548496246338,
1406
+ "eval_runtime": 37.9353,
1407
+ "eval_samples_per_second": 2.214,
1408
+ "eval_steps_per_second": 0.29,
1409
+ "step": 500
1410
+ },
1411
+ {
1412
+ "epoch": 1.3,
1413
+ "learning_rate": 3.508997429305913e-05,
1414
+ "loss": 1.6381,
1415
+ "step": 505
1416
+ },
1417
+ {
1418
+ "epoch": 1.3,
1419
+ "eval_loss": 1.661049246788025,
1420
+ "eval_runtime": 37.9669,
1421
+ "eval_samples_per_second": 2.212,
1422
+ "eval_steps_per_second": 0.29,
1423
+ "step": 505
1424
+ },
1425
+ {
1426
+ "epoch": 1.31,
1427
+ "learning_rate": 3.444730077120823e-05,
1428
+ "loss": 1.6506,
1429
+ "step": 510
1430
+ },
1431
+ {
1432
+ "epoch": 1.31,
1433
+ "eval_loss": 1.6610089540481567,
1434
+ "eval_runtime": 37.9127,
1435
+ "eval_samples_per_second": 2.216,
1436
+ "eval_steps_per_second": 0.29,
1437
+ "step": 510
1438
+ },
1439
+ {
1440
+ "epoch": 1.32,
1441
+ "learning_rate": 3.380462724935733e-05,
1442
+ "loss": 1.6376,
1443
+ "step": 515
1444
+ },
1445
+ {
1446
+ "epoch": 1.32,
1447
+ "eval_loss": 1.6608952283859253,
1448
+ "eval_runtime": 37.9454,
1449
+ "eval_samples_per_second": 2.214,
1450
+ "eval_steps_per_second": 0.29,
1451
+ "step": 515
1452
+ },
1453
+ {
1454
+ "epoch": 1.34,
1455
+ "learning_rate": 3.316195372750643e-05,
1456
+ "loss": 1.6354,
1457
+ "step": 520
1458
+ },
1459
+ {
1460
+ "epoch": 1.34,
1461
+ "eval_loss": 1.6610838174819946,
1462
+ "eval_runtime": 37.7265,
1463
+ "eval_samples_per_second": 2.227,
1464
+ "eval_steps_per_second": 0.292,
1465
+ "step": 520
1466
+ },
1467
+ {
1468
+ "epoch": 1.35,
1469
+ "learning_rate": 3.251928020565553e-05,
1470
+ "loss": 1.6201,
1471
+ "step": 525
1472
+ },
1473
+ {
1474
+ "epoch": 1.35,
1475
+ "eval_loss": 1.6610314846038818,
1476
+ "eval_runtime": 37.8464,
1477
+ "eval_samples_per_second": 2.219,
1478
+ "eval_steps_per_second": 0.291,
1479
+ "step": 525
1480
+ },
1481
+ {
1482
+ "epoch": 1.36,
1483
+ "learning_rate": 3.1876606683804625e-05,
1484
+ "loss": 1.6461,
1485
+ "step": 530
1486
+ },
1487
+ {
1488
+ "epoch": 1.36,
1489
+ "eval_loss": 1.6605802774429321,
1490
+ "eval_runtime": 37.9503,
1491
+ "eval_samples_per_second": 2.213,
1492
+ "eval_steps_per_second": 0.29,
1493
+ "step": 530
1494
+ },
1495
+ {
1496
+ "epoch": 1.38,
1497
+ "learning_rate": 3.1233933161953726e-05,
1498
+ "loss": 1.6818,
1499
+ "step": 535
1500
+ },
1501
+ {
1502
+ "epoch": 1.38,
1503
+ "eval_loss": 1.6607571840286255,
1504
+ "eval_runtime": 37.9086,
1505
+ "eval_samples_per_second": 2.216,
1506
+ "eval_steps_per_second": 0.29,
1507
+ "step": 535
1508
+ },
1509
+ {
1510
+ "epoch": 1.39,
1511
+ "learning_rate": 3.059125964010283e-05,
1512
+ "loss": 1.6117,
1513
+ "step": 540
1514
+ },
1515
+ {
1516
+ "epoch": 1.39,
1517
+ "eval_loss": 1.6605336666107178,
1518
+ "eval_runtime": 37.947,
1519
+ "eval_samples_per_second": 2.214,
1520
+ "eval_steps_per_second": 0.29,
1521
+ "step": 540
1522
+ },
1523
+ {
1524
+ "epoch": 1.4,
1525
+ "learning_rate": 2.994858611825193e-05,
1526
+ "loss": 1.6252,
1527
+ "step": 545
1528
+ },
1529
+ {
1530
+ "epoch": 1.4,
1531
+ "eval_loss": 1.6603213548660278,
1532
+ "eval_runtime": 37.9826,
1533
+ "eval_samples_per_second": 2.212,
1534
+ "eval_steps_per_second": 0.29,
1535
+ "step": 545
1536
+ },
1537
+ {
1538
+ "epoch": 1.41,
1539
+ "learning_rate": 2.930591259640103e-05,
1540
+ "loss": 1.6486,
1541
+ "step": 550
1542
+ },
1543
+ {
1544
+ "epoch": 1.41,
1545
+ "eval_loss": 1.6600297689437866,
1546
+ "eval_runtime": 37.9635,
1547
+ "eval_samples_per_second": 2.213,
1548
+ "eval_steps_per_second": 0.29,
1549
+ "step": 550
1550
+ },
1551
+ {
1552
+ "epoch": 1.43,
1553
+ "learning_rate": 2.866323907455013e-05,
1554
+ "loss": 1.6861,
1555
+ "step": 555
1556
+ },
1557
+ {
1558
+ "epoch": 1.43,
1559
+ "eval_loss": 1.6602742671966553,
1560
+ "eval_runtime": 37.8722,
1561
+ "eval_samples_per_second": 2.218,
1562
+ "eval_steps_per_second": 0.29,
1563
+ "step": 555
1564
+ },
1565
+ {
1566
+ "epoch": 1.44,
1567
+ "learning_rate": 2.802056555269923e-05,
1568
+ "loss": 1.6624,
1569
+ "step": 560
1570
+ },
1571
+ {
1572
+ "epoch": 1.44,
1573
+ "eval_loss": 1.6599992513656616,
1574
+ "eval_runtime": 37.9306,
1575
+ "eval_samples_per_second": 2.215,
1576
+ "eval_steps_per_second": 0.29,
1577
+ "step": 560
1578
+ },
1579
+ {
1580
+ "epoch": 1.45,
1581
+ "learning_rate": 2.737789203084833e-05,
1582
+ "loss": 1.6323,
1583
+ "step": 565
1584
+ },
1585
+ {
1586
+ "epoch": 1.45,
1587
+ "eval_loss": 1.6600306034088135,
1588
+ "eval_runtime": 37.9548,
1589
+ "eval_samples_per_second": 2.213,
1590
+ "eval_steps_per_second": 0.29,
1591
+ "step": 565
1592
+ },
1593
+ {
1594
+ "epoch": 1.47,
1595
+ "learning_rate": 2.673521850899743e-05,
1596
+ "loss": 1.6707,
1597
+ "step": 570
1598
+ },
1599
+ {
1600
+ "epoch": 1.47,
1601
+ "eval_loss": 1.6601014137268066,
1602
+ "eval_runtime": 38.0349,
1603
+ "eval_samples_per_second": 2.208,
1604
+ "eval_steps_per_second": 0.289,
1605
+ "step": 570
1606
+ },
1607
+ {
1608
+ "epoch": 1.48,
1609
+ "learning_rate": 2.6092544987146534e-05,
1610
+ "loss": 1.6478,
1611
+ "step": 575
1612
+ },
1613
+ {
1614
+ "epoch": 1.48,
1615
+ "eval_loss": 1.6597967147827148,
1616
+ "eval_runtime": 37.9413,
1617
+ "eval_samples_per_second": 2.214,
1618
+ "eval_steps_per_second": 0.29,
1619
+ "step": 575
1620
+ },
1621
+ {
1622
+ "epoch": 1.49,
1623
+ "learning_rate": 2.5449871465295634e-05,
1624
+ "loss": 1.6715,
1625
+ "step": 580
1626
+ },
1627
+ {
1628
+ "epoch": 1.49,
1629
+ "eval_loss": 1.6598337888717651,
1630
+ "eval_runtime": 37.8386,
1631
+ "eval_samples_per_second": 2.22,
1632
+ "eval_steps_per_second": 0.291,
1633
+ "step": 580
1634
+ },
1635
+ {
1636
+ "epoch": 1.5,
1637
+ "learning_rate": 2.480719794344473e-05,
1638
+ "loss": 1.6626,
1639
+ "step": 585
1640
+ },
1641
+ {
1642
+ "epoch": 1.5,
1643
+ "eval_loss": 1.6600011587142944,
1644
+ "eval_runtime": 37.7901,
1645
+ "eval_samples_per_second": 2.223,
1646
+ "eval_steps_per_second": 0.291,
1647
+ "step": 585
1648
+ },
1649
+ {
1650
+ "epoch": 1.52,
1651
+ "learning_rate": 2.4164524421593832e-05,
1652
+ "loss": 1.651,
1653
+ "step": 590
1654
+ },
1655
+ {
1656
+ "epoch": 1.52,
1657
+ "eval_loss": 1.65969717502594,
1658
+ "eval_runtime": 37.7643,
1659
+ "eval_samples_per_second": 2.224,
1660
+ "eval_steps_per_second": 0.291,
1661
+ "step": 590
1662
+ },
1663
+ {
1664
+ "epoch": 1.53,
1665
+ "learning_rate": 2.3521850899742933e-05,
1666
+ "loss": 1.6346,
1667
+ "step": 595
1668
+ },
1669
+ {
1670
+ "epoch": 1.53,
1671
+ "eval_loss": 1.6598676443099976,
1672
+ "eval_runtime": 37.9026,
1673
+ "eval_samples_per_second": 2.216,
1674
+ "eval_steps_per_second": 0.29,
1675
+ "step": 595
1676
+ },
1677
+ {
1678
+ "epoch": 1.54,
1679
+ "learning_rate": 2.2879177377892033e-05,
1680
+ "loss": 1.6533,
1681
+ "step": 600
1682
+ },
1683
+ {
1684
+ "epoch": 1.54,
1685
+ "eval_loss": 1.659874677658081,
1686
+ "eval_runtime": 37.9871,
1687
+ "eval_samples_per_second": 2.211,
1688
+ "eval_steps_per_second": 0.29,
1689
+ "step": 600
1690
+ },
1691
+ {
1692
+ "epoch": 1.56,
1693
+ "learning_rate": 2.2236503856041134e-05,
1694
+ "loss": 1.6913,
1695
+ "step": 605
1696
+ },
1697
+ {
1698
+ "epoch": 1.56,
1699
+ "eval_loss": 1.659849762916565,
1700
+ "eval_runtime": 37.8712,
1701
+ "eval_samples_per_second": 2.218,
1702
+ "eval_steps_per_second": 0.29,
1703
+ "step": 605
1704
+ },
1705
+ {
1706
+ "epoch": 1.57,
1707
+ "learning_rate": 2.159383033419023e-05,
1708
+ "loss": 1.6709,
1709
+ "step": 610
1710
+ },
1711
+ {
1712
+ "epoch": 1.57,
1713
+ "eval_loss": 1.6598856449127197,
1714
+ "eval_runtime": 37.7043,
1715
+ "eval_samples_per_second": 2.228,
1716
+ "eval_steps_per_second": 0.292,
1717
+ "step": 610
1718
+ },
1719
+ {
1720
+ "epoch": 1.58,
1721
+ "learning_rate": 2.095115681233933e-05,
1722
+ "loss": 1.598,
1723
+ "step": 615
1724
+ },
1725
+ {
1726
+ "epoch": 1.58,
1727
+ "eval_loss": 1.6597983837127686,
1728
+ "eval_runtime": 37.722,
1729
+ "eval_samples_per_second": 2.227,
1730
+ "eval_steps_per_second": 0.292,
1731
+ "step": 615
1732
+ },
1733
+ {
1734
+ "epoch": 1.59,
1735
+ "learning_rate": 2.0308483290488432e-05,
1736
+ "loss": 1.6434,
1737
+ "step": 620
1738
+ },
1739
+ {
1740
+ "epoch": 1.59,
1741
+ "eval_loss": 1.6595433950424194,
1742
+ "eval_runtime": 37.6125,
1743
+ "eval_samples_per_second": 2.233,
1744
+ "eval_steps_per_second": 0.292,
1745
+ "step": 620
1746
+ },
1747
+ {
1748
+ "epoch": 1.61,
1749
+ "learning_rate": 1.9665809768637533e-05,
1750
+ "loss": 1.6086,
1751
+ "step": 625
1752
+ },
1753
+ {
1754
+ "epoch": 1.61,
1755
+ "eval_loss": 1.6594945192337036,
1756
+ "eval_runtime": 38.0046,
1757
+ "eval_samples_per_second": 2.21,
1758
+ "eval_steps_per_second": 0.289,
1759
+ "step": 625
1760
+ },
1761
+ {
1762
+ "epoch": 1.62,
1763
+ "learning_rate": 1.9023136246786633e-05,
1764
+ "loss": 1.5962,
1765
+ "step": 630
1766
+ },
1767
+ {
1768
+ "epoch": 1.62,
1769
+ "eval_loss": 1.6595605611801147,
1770
+ "eval_runtime": 37.8518,
1771
+ "eval_samples_per_second": 2.219,
1772
+ "eval_steps_per_second": 0.291,
1773
+ "step": 630
1774
+ },
1775
+ {
1776
+ "epoch": 1.63,
1777
+ "learning_rate": 1.8380462724935734e-05,
1778
+ "loss": 1.6695,
1779
+ "step": 635
1780
+ },
1781
+ {
1782
+ "epoch": 1.63,
1783
+ "eval_loss": 1.6592403650283813,
1784
+ "eval_runtime": 37.7285,
1785
+ "eval_samples_per_second": 2.226,
1786
+ "eval_steps_per_second": 0.292,
1787
+ "step": 635
1788
+ },
1789
+ {
1790
+ "epoch": 1.65,
1791
+ "learning_rate": 1.7737789203084834e-05,
1792
+ "loss": 1.6526,
1793
+ "step": 640
1794
+ },
1795
+ {
1796
+ "epoch": 1.65,
1797
+ "eval_loss": 1.6591880321502686,
1798
+ "eval_runtime": 37.758,
1799
+ "eval_samples_per_second": 2.225,
1800
+ "eval_steps_per_second": 0.291,
1801
+ "step": 640
1802
+ },
1803
+ {
1804
+ "epoch": 1.66,
1805
+ "learning_rate": 1.7095115681233935e-05,
1806
+ "loss": 1.672,
1807
+ "step": 645
1808
+ },
1809
+ {
1810
+ "epoch": 1.66,
1811
+ "eval_loss": 1.6588138341903687,
1812
+ "eval_runtime": 37.92,
1813
+ "eval_samples_per_second": 2.215,
1814
+ "eval_steps_per_second": 0.29,
1815
+ "step": 645
1816
+ },
1817
+ {
1818
+ "epoch": 1.67,
1819
+ "learning_rate": 1.6452442159383032e-05,
1820
+ "loss": 1.6347,
1821
+ "step": 650
1822
+ },
1823
+ {
1824
+ "epoch": 1.67,
1825
+ "eval_loss": 1.6589809656143188,
1826
+ "eval_runtime": 37.8021,
1827
+ "eval_samples_per_second": 2.222,
1828
+ "eval_steps_per_second": 0.291,
1829
+ "step": 650
1830
+ },
1831
+ {
1832
+ "epoch": 1.68,
1833
+ "learning_rate": 1.5809768637532136e-05,
1834
+ "loss": 1.6227,
1835
+ "step": 655
1836
+ },
1837
+ {
1838
+ "epoch": 1.68,
1839
+ "eval_loss": 1.6586792469024658,
1840
+ "eval_runtime": 37.9128,
1841
+ "eval_samples_per_second": 2.216,
1842
+ "eval_steps_per_second": 0.29,
1843
+ "step": 655
1844
+ },
1845
+ {
1846
+ "epoch": 1.7,
1847
+ "learning_rate": 1.5167095115681235e-05,
1848
+ "loss": 1.6308,
1849
+ "step": 660
1850
+ },
1851
+ {
1852
+ "epoch": 1.7,
1853
+ "eval_loss": 1.6585420370101929,
1854
+ "eval_runtime": 37.9881,
1855
+ "eval_samples_per_second": 2.211,
1856
+ "eval_steps_per_second": 0.29,
1857
+ "step": 660
1858
+ },
1859
+ {
1860
+ "epoch": 1.71,
1861
+ "learning_rate": 1.4524421593830334e-05,
1862
+ "loss": 1.6495,
1863
+ "step": 665
1864
+ },
1865
+ {
1866
+ "epoch": 1.71,
1867
+ "eval_loss": 1.6587530374526978,
1868
+ "eval_runtime": 37.9148,
1869
+ "eval_samples_per_second": 2.215,
1870
+ "eval_steps_per_second": 0.29,
1871
+ "step": 665
1872
+ },
1873
+ {
1874
+ "epoch": 1.72,
1875
+ "learning_rate": 1.3881748071979436e-05,
1876
+ "loss": 1.6712,
1877
+ "step": 670
1878
+ },
1879
+ {
1880
+ "epoch": 1.72,
1881
+ "eval_loss": 1.658648133277893,
1882
+ "eval_runtime": 37.9313,
1883
+ "eval_samples_per_second": 2.215,
1884
+ "eval_steps_per_second": 0.29,
1885
+ "step": 670
1886
+ },
1887
+ {
1888
+ "epoch": 1.74,
1889
+ "learning_rate": 1.3239074550128535e-05,
1890
+ "loss": 1.636,
1891
+ "step": 675
1892
+ },
1893
+ {
1894
+ "epoch": 1.74,
1895
+ "eval_loss": 1.6584407091140747,
1896
+ "eval_runtime": 37.7963,
1897
+ "eval_samples_per_second": 2.222,
1898
+ "eval_steps_per_second": 0.291,
1899
+ "step": 675
1900
+ },
1901
+ {
1902
+ "epoch": 1.75,
1903
+ "learning_rate": 1.2596401028277636e-05,
1904
+ "loss": 1.6453,
1905
+ "step": 680
1906
+ },
1907
+ {
1908
+ "epoch": 1.75,
1909
+ "eval_loss": 1.658594012260437,
1910
+ "eval_runtime": 37.7226,
1911
+ "eval_samples_per_second": 2.227,
1912
+ "eval_steps_per_second": 0.292,
1913
+ "step": 680
1914
+ },
1915
+ {
1916
+ "epoch": 1.76,
1917
+ "learning_rate": 1.1953727506426736e-05,
1918
+ "loss": 1.6509,
1919
+ "step": 685
1920
+ },
1921
+ {
1922
+ "epoch": 1.76,
1923
+ "eval_loss": 1.658467411994934,
1924
+ "eval_runtime": 37.8179,
1925
+ "eval_samples_per_second": 2.221,
1926
+ "eval_steps_per_second": 0.291,
1927
+ "step": 685
1928
+ },
1929
+ {
1930
+ "epoch": 1.77,
1931
+ "learning_rate": 1.1311053984575835e-05,
1932
+ "loss": 1.6489,
1933
+ "step": 690
1934
+ },
1935
+ {
1936
+ "epoch": 1.77,
1937
+ "eval_loss": 1.6585352420806885,
1938
+ "eval_runtime": 37.6768,
1939
+ "eval_samples_per_second": 2.229,
1940
+ "eval_steps_per_second": 0.292,
1941
+ "step": 690
1942
+ },
1943
+ {
1944
+ "epoch": 1.79,
1945
+ "learning_rate": 1.0668380462724936e-05,
1946
+ "loss": 1.6424,
1947
+ "step": 695
1948
+ },
1949
+ {
1950
+ "epoch": 1.79,
1951
+ "eval_loss": 1.6582835912704468,
1952
+ "eval_runtime": 38.0083,
1953
+ "eval_samples_per_second": 2.21,
1954
+ "eval_steps_per_second": 0.289,
1955
+ "step": 695
1956
+ },
1957
+ {
1958
+ "epoch": 1.8,
1959
+ "learning_rate": 1.0025706940874038e-05,
1960
+ "loss": 1.6131,
1961
+ "step": 700
1962
+ },
1963
+ {
1964
+ "epoch": 1.8,
1965
+ "eval_loss": 1.658105731010437,
1966
+ "eval_runtime": 37.8711,
1967
+ "eval_samples_per_second": 2.218,
1968
+ "eval_steps_per_second": 0.29,
1969
+ "step": 700
1970
+ },
1971
+ {
1972
+ "epoch": 1.81,
1973
+ "learning_rate": 9.383033419023137e-06,
1974
+ "loss": 1.6225,
1975
+ "step": 705
1976
+ },
1977
+ {
1978
+ "epoch": 1.81,
1979
+ "eval_loss": 1.658282995223999,
1980
+ "eval_runtime": 37.8125,
1981
+ "eval_samples_per_second": 2.221,
1982
+ "eval_steps_per_second": 0.291,
1983
+ "step": 705
1984
+ },
1985
+ {
1986
+ "epoch": 1.83,
1987
+ "learning_rate": 8.740359897172237e-06,
1988
+ "loss": 1.6928,
1989
+ "step": 710
1990
+ },
1991
+ {
1992
+ "epoch": 1.83,
1993
+ "eval_loss": 1.6581056118011475,
1994
+ "eval_runtime": 37.863,
1995
+ "eval_samples_per_second": 2.219,
1996
+ "eval_steps_per_second": 0.291,
1997
+ "step": 710
1998
+ },
1999
+ {
2000
+ "epoch": 1.84,
2001
+ "learning_rate": 8.097686375321336e-06,
2002
+ "loss": 1.6227,
2003
+ "step": 715
2004
+ },
2005
+ {
2006
+ "epoch": 1.84,
2007
+ "eval_loss": 1.6581188440322876,
2008
+ "eval_runtime": 37.8646,
2009
+ "eval_samples_per_second": 2.218,
2010
+ "eval_steps_per_second": 0.291,
2011
+ "step": 715
2012
+ },
2013
+ {
2014
+ "epoch": 1.85,
2015
+ "learning_rate": 7.4550128534704376e-06,
2016
+ "loss": 1.6593,
2017
+ "step": 720
2018
+ },
2019
+ {
2020
+ "epoch": 1.85,
2021
+ "eval_loss": 1.6579999923706055,
2022
+ "eval_runtime": 37.876,
2023
+ "eval_samples_per_second": 2.218,
2024
+ "eval_steps_per_second": 0.29,
2025
+ "step": 720
2026
+ },
2027
+ {
2028
+ "epoch": 1.86,
2029
+ "learning_rate": 6.812339331619537e-06,
2030
+ "loss": 1.6642,
2031
+ "step": 725
2032
+ },
2033
+ {
2034
+ "epoch": 1.86,
2035
+ "eval_loss": 1.65813410282135,
2036
+ "eval_runtime": 37.7093,
2037
+ "eval_samples_per_second": 2.228,
2038
+ "eval_steps_per_second": 0.292,
2039
+ "step": 725
2040
+ },
2041
+ {
2042
+ "epoch": 1.88,
2043
+ "learning_rate": 6.169665809768638e-06,
2044
+ "loss": 1.6414,
2045
+ "step": 730
2046
+ },
2047
+ {
2048
+ "epoch": 1.88,
2049
+ "eval_loss": 1.6581989526748657,
2050
+ "eval_runtime": 37.7818,
2051
+ "eval_samples_per_second": 2.223,
2052
+ "eval_steps_per_second": 0.291,
2053
+ "step": 730
2054
+ },
2055
+ {
2056
+ "epoch": 1.89,
2057
+ "learning_rate": 5.526992287917738e-06,
2058
+ "loss": 1.6519,
2059
+ "step": 735
2060
+ },
2061
+ {
2062
+ "epoch": 1.89,
2063
+ "eval_loss": 1.6577394008636475,
2064
+ "eval_runtime": 37.7264,
2065
+ "eval_samples_per_second": 2.227,
2066
+ "eval_steps_per_second": 0.292,
2067
+ "step": 735
2068
+ },
2069
+ {
2070
+ "epoch": 1.9,
2071
+ "learning_rate": 4.884318766066838e-06,
2072
+ "loss": 1.6485,
2073
+ "step": 740
2074
+ },
2075
+ {
2076
+ "epoch": 1.9,
2077
+ "eval_loss": 1.6580238342285156,
2078
+ "eval_runtime": 37.8729,
2079
+ "eval_samples_per_second": 2.218,
2080
+ "eval_steps_per_second": 0.29,
2081
+ "step": 740
2082
+ },
2083
+ {
2084
+ "epoch": 1.92,
2085
+ "learning_rate": 4.241645244215939e-06,
2086
+ "loss": 1.6261,
2087
+ "step": 745
2088
+ },
2089
+ {
2090
+ "epoch": 1.92,
2091
+ "eval_loss": 1.6580148935317993,
2092
+ "eval_runtime": 37.9546,
2093
+ "eval_samples_per_second": 2.213,
2094
+ "eval_steps_per_second": 0.29,
2095
+ "step": 745
2096
+ },
2097
+ {
2098
+ "epoch": 1.93,
2099
+ "learning_rate": 3.598971722365039e-06,
2100
+ "loss": 1.657,
2101
+ "step": 750
2102
+ },
2103
+ {
2104
+ "epoch": 1.93,
2105
+ "eval_loss": 1.657867670059204,
2106
+ "eval_runtime": 37.9411,
2107
+ "eval_samples_per_second": 2.214,
2108
+ "eval_steps_per_second": 0.29,
2109
+ "step": 750
2110
+ },
2111
+ {
2112
+ "epoch": 1.94,
2113
+ "learning_rate": 2.956298200514139e-06,
2114
+ "loss": 1.6535,
2115
+ "step": 755
2116
+ },
2117
+ {
2118
+ "epoch": 1.94,
2119
+ "eval_loss": 1.6581624746322632,
2120
+ "eval_runtime": 37.8532,
2121
+ "eval_samples_per_second": 2.219,
2122
+ "eval_steps_per_second": 0.291,
2123
+ "step": 755
2124
+ },
2125
+ {
2126
+ "epoch": 1.95,
2127
+ "learning_rate": 2.313624678663239e-06,
2128
+ "loss": 1.6518,
2129
+ "step": 760
2130
+ },
2131
+ {
2132
+ "epoch": 1.95,
2133
+ "eval_loss": 1.6577980518341064,
2134
+ "eval_runtime": 37.657,
2135
+ "eval_samples_per_second": 2.231,
2136
+ "eval_steps_per_second": 0.292,
2137
+ "step": 760
2138
+ },
2139
+ {
2140
+ "epoch": 1.97,
2141
+ "learning_rate": 1.6709511568123394e-06,
2142
+ "loss": 1.6946,
2143
+ "step": 765
2144
+ },
2145
+ {
2146
+ "epoch": 1.97,
2147
+ "eval_loss": 1.6577955484390259,
2148
+ "eval_runtime": 37.7796,
2149
+ "eval_samples_per_second": 2.223,
2150
+ "eval_steps_per_second": 0.291,
2151
+ "step": 765
2152
+ },
2153
+ {
2154
+ "epoch": 1.98,
2155
+ "learning_rate": 1.0282776349614395e-06,
2156
+ "loss": 1.6325,
2157
+ "step": 770
2158
+ },
2159
+ {
2160
+ "epoch": 1.98,
2161
+ "eval_loss": 1.6578012704849243,
2162
+ "eval_runtime": 37.71,
2163
+ "eval_samples_per_second": 2.228,
2164
+ "eval_steps_per_second": 0.292,
2165
+ "step": 770
2166
+ }
2167
+ ],
2168
+ "logging_steps": 5,
2169
+ "max_steps": 778,
2170
+ "num_train_epochs": 2,
2171
+ "save_steps": 10,
2172
+ "total_flos": 2.5014589754179584e+17,
2173
+ "trial_name": null,
2174
+ "trial_params": null
2175
+ }
checkpoint-770/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c6211dd8a7344cda681d69cc75786fde8b7049338b76992777b83de12eab260a
3
+ size 4027
config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "meta-llama/Llama-2-7b-hf",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "eos_token_id": 2,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 4096,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 11008,
12
+ "max_position_embeddings": 4096,
13
+ "model_type": "llama",
14
+ "num_attention_heads": 32,
15
+ "num_hidden_layers": 32,
16
+ "num_key_value_heads": 32,
17
+ "pretraining_tp": 1,
18
+ "quantization_config": {
19
+ "bnb_4bit_compute_dtype": "float32",
20
+ "bnb_4bit_quant_type": "fp4",
21
+ "bnb_4bit_use_double_quant": false,
22
+ "llm_int8_enable_fp32_cpu_offload": false,
23
+ "llm_int8_has_fp16_weight": false,
24
+ "llm_int8_skip_modules": null,
25
+ "llm_int8_threshold": 6.0,
26
+ "load_in_4bit": false,
27
+ "load_in_8bit": true,
28
+ "quant_method": "bitsandbytes"
29
+ },
30
+ "rms_norm_eps": 1e-05,
31
+ "rope_scaling": null,
32
+ "tie_word_embeddings": false,
33
+ "torch_dtype": "float16",
34
+ "transformers_version": "4.32.1",
35
+ "use_cache": true,
36
+ "vocab_size": 32000
37
+ }
peft_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "default": {
3
+ "peft_type": "LORA",
4
+ "auto_mapping": null,
5
+ "base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
6
+ "revision": null,
7
+ "task_type": "CAUSAL_LM",
8
+ "inference_mode": false,
9
+ "r": 8,
10
+ "target_modules": [
11
+ "q_proj",
12
+ "v_proj"
13
+ ],
14
+ "lora_alpha": 32,
15
+ "lora_dropout": 0.05,
16
+ "fan_in_fan_out": false,
17
+ "bias": "none",
18
+ "modules_to_save": null,
19
+ "init_lora_weights": true,
20
+ "layers_to_transform": null,
21
+ "layers_pattern": null
22
+ }
23
+ }