End of training
Browse files- README.md +2 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +1095 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
@@ -4,6 +4,7 @@ license: apache-2.0
|
|
4 |
base_model: mistralai/Mistral-7B-v0.3
|
5 |
tags:
|
6 |
- llama-factory
|
|
|
7 |
- generated_from_trainer
|
8 |
model-index:
|
9 |
- name: mistral_7b_0-3_oh-dcft-v3.1-llama-3.1-405b
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
# mistral_7b_0-3_oh-dcft-v3.1-llama-3.1-405b
|
17 |
|
18 |
-
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on
|
19 |
It achieves the following results on the evaluation set:
|
20 |
- Loss: 0.2729
|
21 |
|
|
|
4 |
base_model: mistralai/Mistral-7B-v0.3
|
5 |
tags:
|
6 |
- llama-factory
|
7 |
+
- full
|
8 |
- generated_from_trainer
|
9 |
model-index:
|
10 |
- name: mistral_7b_0-3_oh-dcft-v3.1-llama-3.1-405b
|
|
|
16 |
|
17 |
# mistral_7b_0-3_oh-dcft-v3.1-llama-3.1-405b
|
18 |
|
19 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the mlfoundations-dev/oh-dcft-v3.1-llama-3.1-405b dataset.
|
20 |
It achieves the following results on the evaluation set:
|
21 |
- Loss: 0.2729
|
22 |
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_loss": 0.2728930115699768,
|
4 |
+
"eval_runtime": 43.9016,
|
5 |
+
"eval_samples_per_second": 301.15,
|
6 |
+
"eval_steps_per_second": 1.184,
|
7 |
+
"total_flos": 2467329862533120.0,
|
8 |
+
"train_loss": 0.19335093406374892,
|
9 |
+
"train_runtime": 8782.5965,
|
10 |
+
"train_samples_per_second": 85.801,
|
11 |
+
"train_steps_per_second": 0.168
|
12 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_loss": 0.2728930115699768,
|
4 |
+
"eval_runtime": 43.9016,
|
5 |
+
"eval_samples_per_second": 301.15,
|
6 |
+
"eval_steps_per_second": 1.184
|
7 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"total_flos": 2467329862533120.0,
|
4 |
+
"train_loss": 0.19335093406374892,
|
5 |
+
"train_runtime": 8782.5965,
|
6 |
+
"train_samples_per_second": 85.801,
|
7 |
+
"train_steps_per_second": 0.168
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,1095 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 3.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1473,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.020366598778004074,
|
13 |
+
"grad_norm": 7.34504383318826,
|
14 |
+
"learning_rate": 6.756756756756758e-07,
|
15 |
+
"loss": 0.5197,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.04073319755600815,
|
20 |
+
"grad_norm": 1.9502076909187314,
|
21 |
+
"learning_rate": 1.3513513513513515e-06,
|
22 |
+
"loss": 0.4425,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.06109979633401222,
|
27 |
+
"grad_norm": 1.3703382365758696,
|
28 |
+
"learning_rate": 2.0270270270270273e-06,
|
29 |
+
"loss": 0.3956,
|
30 |
+
"step": 30
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.0814663951120163,
|
34 |
+
"grad_norm": 1.5274033164070318,
|
35 |
+
"learning_rate": 2.702702702702703e-06,
|
36 |
+
"loss": 0.3706,
|
37 |
+
"step": 40
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.10183299389002037,
|
41 |
+
"grad_norm": 1.5387129397990396,
|
42 |
+
"learning_rate": 3.3783783783783788e-06,
|
43 |
+
"loss": 0.3558,
|
44 |
+
"step": 50
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.12219959266802444,
|
48 |
+
"grad_norm": 1.5285386592148773,
|
49 |
+
"learning_rate": 4.0540540540540545e-06,
|
50 |
+
"loss": 0.3439,
|
51 |
+
"step": 60
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.1425661914460285,
|
55 |
+
"grad_norm": 1.4343248038161889,
|
56 |
+
"learning_rate": 4.72972972972973e-06,
|
57 |
+
"loss": 0.3351,
|
58 |
+
"step": 70
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.1629327902240326,
|
62 |
+
"grad_norm": 1.3758371691545166,
|
63 |
+
"learning_rate": 4.999795773182583e-06,
|
64 |
+
"loss": 0.3273,
|
65 |
+
"step": 80
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.18329938900203666,
|
69 |
+
"grad_norm": 1.5978694880041386,
|
70 |
+
"learning_rate": 4.998547854667808e-06,
|
71 |
+
"loss": 0.3217,
|
72 |
+
"step": 90
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.20366598778004075,
|
76 |
+
"grad_norm": 1.3851731195685373,
|
77 |
+
"learning_rate": 4.9961661054815474e-06,
|
78 |
+
"loss": 0.322,
|
79 |
+
"step": 100
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.2240325865580448,
|
83 |
+
"grad_norm": 1.3521127973831295,
|
84 |
+
"learning_rate": 4.992651726621268e-06,
|
85 |
+
"loss": 0.3146,
|
86 |
+
"step": 110
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.24439918533604887,
|
90 |
+
"grad_norm": 1.3516467751032153,
|
91 |
+
"learning_rate": 4.988006490213162e-06,
|
92 |
+
"loss": 0.3143,
|
93 |
+
"step": 120
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.26476578411405294,
|
97 |
+
"grad_norm": 1.2701037565844242,
|
98 |
+
"learning_rate": 4.9822327386185436e-06,
|
99 |
+
"loss": 0.3077,
|
100 |
+
"step": 130
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.285132382892057,
|
104 |
+
"grad_norm": 1.6979740720879781,
|
105 |
+
"learning_rate": 4.975333383252724e-06,
|
106 |
+
"loss": 0.307,
|
107 |
+
"step": 140
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.3054989816700611,
|
111 |
+
"grad_norm": 1.3490436638938983,
|
112 |
+
"learning_rate": 4.967311903116924e-06,
|
113 |
+
"loss": 0.2987,
|
114 |
+
"step": 150
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.3258655804480652,
|
118 |
+
"grad_norm": 1.7161350768628911,
|
119 |
+
"learning_rate": 4.958172343043982e-06,
|
120 |
+
"loss": 0.2999,
|
121 |
+
"step": 160
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.34623217922606925,
|
125 |
+
"grad_norm": 1.382612250934794,
|
126 |
+
"learning_rate": 4.947919311658756e-06,
|
127 |
+
"loss": 0.2991,
|
128 |
+
"step": 170
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.3665987780040733,
|
132 |
+
"grad_norm": 1.3151580723383025,
|
133 |
+
"learning_rate": 4.936557979054219e-06,
|
134 |
+
"loss": 0.2949,
|
135 |
+
"step": 180
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.3869653767820774,
|
139 |
+
"grad_norm": 1.3980284114864507,
|
140 |
+
"learning_rate": 4.924094074184436e-06,
|
141 |
+
"loss": 0.2931,
|
142 |
+
"step": 190
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.4073319755600815,
|
146 |
+
"grad_norm": 1.4995260622729383,
|
147 |
+
"learning_rate": 4.910533881975742e-06,
|
148 |
+
"loss": 0.2897,
|
149 |
+
"step": 200
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.42769857433808556,
|
153 |
+
"grad_norm": 1.3091750919874185,
|
154 |
+
"learning_rate": 4.895884240157568e-06,
|
155 |
+
"loss": 0.2897,
|
156 |
+
"step": 210
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.4480651731160896,
|
160 |
+
"grad_norm": 1.4747628264164012,
|
161 |
+
"learning_rate": 4.880152535814507e-06,
|
162 |
+
"loss": 0.2868,
|
163 |
+
"step": 220
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.4684317718940937,
|
167 |
+
"grad_norm": 1.4470008213149266,
|
168 |
+
"learning_rate": 4.863346701661384e-06,
|
169 |
+
"loss": 0.2874,
|
170 |
+
"step": 230
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.48879837067209775,
|
174 |
+
"grad_norm": 1.4515164854269305,
|
175 |
+
"learning_rate": 4.845475212043178e-06,
|
176 |
+
"loss": 0.2878,
|
177 |
+
"step": 240
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.5091649694501018,
|
181 |
+
"grad_norm": 1.2578053743674917,
|
182 |
+
"learning_rate": 4.8265470786618345e-06,
|
183 |
+
"loss": 0.2811,
|
184 |
+
"step": 250
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.5295315682281059,
|
188 |
+
"grad_norm": 1.5098983649320945,
|
189 |
+
"learning_rate": 4.806571846032109e-06,
|
190 |
+
"loss": 0.2773,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.5498981670061099,
|
195 |
+
"grad_norm": 1.260736734336587,
|
196 |
+
"learning_rate": 4.785559586668751e-06,
|
197 |
+
"loss": 0.28,
|
198 |
+
"step": 270
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.570264765784114,
|
202 |
+
"grad_norm": 1.1734721692297405,
|
203 |
+
"learning_rate": 4.763520896007427e-06,
|
204 |
+
"loss": 0.2825,
|
205 |
+
"step": 280
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.5906313645621182,
|
209 |
+
"grad_norm": 1.1031919455840085,
|
210 |
+
"learning_rate": 4.74046688706198e-06,
|
211 |
+
"loss": 0.2759,
|
212 |
+
"step": 290
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.6109979633401222,
|
216 |
+
"grad_norm": 1.0944761517106407,
|
217 |
+
"learning_rate": 4.716409184820684e-06,
|
218 |
+
"loss": 0.2766,
|
219 |
+
"step": 300
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.6313645621181263,
|
223 |
+
"grad_norm": 1.23087624035179,
|
224 |
+
"learning_rate": 4.691359920384341e-06,
|
225 |
+
"loss": 0.2777,
|
226 |
+
"step": 310
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.6517311608961304,
|
230 |
+
"grad_norm": 1.1545647683752378,
|
231 |
+
"learning_rate": 4.6653317248491754e-06,
|
232 |
+
"loss": 0.2752,
|
233 |
+
"step": 320
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.6720977596741344,
|
237 |
+
"grad_norm": 1.2931986965229072,
|
238 |
+
"learning_rate": 4.638337722937599e-06,
|
239 |
+
"loss": 0.2746,
|
240 |
+
"step": 330
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.6924643584521385,
|
244 |
+
"grad_norm": 1.1273344784116983,
|
245 |
+
"learning_rate": 4.610391526380066e-06,
|
246 |
+
"loss": 0.2688,
|
247 |
+
"step": 340
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.7128309572301426,
|
251 |
+
"grad_norm": 1.415362493265054,
|
252 |
+
"learning_rate": 4.581507227051359e-06,
|
253 |
+
"loss": 0.2703,
|
254 |
+
"step": 350
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.7331975560081466,
|
258 |
+
"grad_norm": 1.1591318242459208,
|
259 |
+
"learning_rate": 4.551699389864756e-06,
|
260 |
+
"loss": 0.2665,
|
261 |
+
"step": 360
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.7535641547861507,
|
265 |
+
"grad_norm": 1.0554558469893176,
|
266 |
+
"learning_rate": 4.520983045427667e-06,
|
267 |
+
"loss": 0.2693,
|
268 |
+
"step": 370
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.7739307535641547,
|
272 |
+
"grad_norm": 1.2265072577256237,
|
273 |
+
"learning_rate": 4.489373682462457e-06,
|
274 |
+
"loss": 0.2663,
|
275 |
+
"step": 380
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.7942973523421588,
|
279 |
+
"grad_norm": 1.2855575031097177,
|
280 |
+
"learning_rate": 4.456887239996245e-06,
|
281 |
+
"loss": 0.2679,
|
282 |
+
"step": 390
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.814663951120163,
|
286 |
+
"grad_norm": 1.081374466408974,
|
287 |
+
"learning_rate": 4.423540099323651e-06,
|
288 |
+
"loss": 0.2661,
|
289 |
+
"step": 400
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.835030549898167,
|
293 |
+
"grad_norm": 1.1990325052929929,
|
294 |
+
"learning_rate": 4.3893490757465264e-06,
|
295 |
+
"loss": 0.2648,
|
296 |
+
"step": 410
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.8553971486761711,
|
300 |
+
"grad_norm": 1.2726034950554865,
|
301 |
+
"learning_rate": 4.354331410094831e-06,
|
302 |
+
"loss": 0.2659,
|
303 |
+
"step": 420
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.8757637474541752,
|
307 |
+
"grad_norm": 1.1644285218921815,
|
308 |
+
"learning_rate": 4.318504760032946e-06,
|
309 |
+
"loss": 0.2613,
|
310 |
+
"step": 430
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.8961303462321792,
|
314 |
+
"grad_norm": 1.236076922309445,
|
315 |
+
"learning_rate": 4.281887191155788e-06,
|
316 |
+
"loss": 0.2636,
|
317 |
+
"step": 440
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.9164969450101833,
|
321 |
+
"grad_norm": 1.1721589750974497,
|
322 |
+
"learning_rate": 4.244497167879234e-06,
|
323 |
+
"loss": 0.261,
|
324 |
+
"step": 450
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.9368635437881874,
|
328 |
+
"grad_norm": 1.085539973119722,
|
329 |
+
"learning_rate": 4.206353544129437e-06,
|
330 |
+
"loss": 0.2632,
|
331 |
+
"step": 460
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.9572301425661914,
|
335 |
+
"grad_norm": 1.1393530942547674,
|
336 |
+
"learning_rate": 4.1674755538357234e-06,
|
337 |
+
"loss": 0.2593,
|
338 |
+
"step": 470
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.9775967413441955,
|
342 |
+
"grad_norm": 1.144980396038313,
|
343 |
+
"learning_rate": 4.12788280123189e-06,
|
344 |
+
"loss": 0.2582,
|
345 |
+
"step": 480
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.9979633401221996,
|
349 |
+
"grad_norm": 1.1413918380162131,
|
350 |
+
"learning_rate": 4.087595250970756e-06,
|
351 |
+
"loss": 0.2555,
|
352 |
+
"step": 490
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 1.0,
|
356 |
+
"eval_loss": 0.2573759853839874,
|
357 |
+
"eval_runtime": 44.646,
|
358 |
+
"eval_samples_per_second": 296.13,
|
359 |
+
"eval_steps_per_second": 1.165,
|
360 |
+
"step": 491
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 1.0183299389002036,
|
364 |
+
"grad_norm": 1.5557280500135056,
|
365 |
+
"learning_rate": 4.0466332180569904e-06,
|
366 |
+
"loss": 0.1936,
|
367 |
+
"step": 500
|
368 |
+
},
|
369 |
+
{
|
370 |
+
"epoch": 1.0386965376782078,
|
371 |
+
"grad_norm": 1.3238755177486623,
|
372 |
+
"learning_rate": 4.0050173576032644e-06,
|
373 |
+
"loss": 0.1814,
|
374 |
+
"step": 510
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 1.0590631364562118,
|
378 |
+
"grad_norm": 1.2066120698543659,
|
379 |
+
"learning_rate": 3.9627686544149085e-06,
|
380 |
+
"loss": 0.181,
|
381 |
+
"step": 520
|
382 |
+
},
|
383 |
+
{
|
384 |
+
"epoch": 1.079429735234216,
|
385 |
+
"grad_norm": 1.088083679916152,
|
386 |
+
"learning_rate": 3.919908412408315e-06,
|
387 |
+
"loss": 0.1799,
|
388 |
+
"step": 530
|
389 |
+
},
|
390 |
+
{
|
391 |
+
"epoch": 1.0997963340122199,
|
392 |
+
"grad_norm": 1.123343825207487,
|
393 |
+
"learning_rate": 3.876458243868441e-06,
|
394 |
+
"loss": 0.1813,
|
395 |
+
"step": 540
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 1.120162932790224,
|
399 |
+
"grad_norm": 1.3201033277961134,
|
400 |
+
"learning_rate": 3.832440058550809e-06,
|
401 |
+
"loss": 0.1815,
|
402 |
+
"step": 550
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 1.140529531568228,
|
406 |
+
"grad_norm": 1.278395529927195,
|
407 |
+
"learning_rate": 3.7878760526335024e-06,
|
408 |
+
"loss": 0.1781,
|
409 |
+
"step": 560
|
410 |
+
},
|
411 |
+
{
|
412 |
+
"epoch": 1.1608961303462322,
|
413 |
+
"grad_norm": 1.11604395104044,
|
414 |
+
"learning_rate": 3.742788697524746e-06,
|
415 |
+
"loss": 0.1773,
|
416 |
+
"step": 570
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 1.1812627291242364,
|
420 |
+
"grad_norm": 1.1089540254153334,
|
421 |
+
"learning_rate": 3.6972007285316907e-06,
|
422 |
+
"loss": 0.1806,
|
423 |
+
"step": 580
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 1.2016293279022403,
|
427 |
+
"grad_norm": 1.1472218678631292,
|
428 |
+
"learning_rate": 3.6511351333961297e-06,
|
429 |
+
"loss": 0.1795,
|
430 |
+
"step": 590
|
431 |
+
},
|
432 |
+
{
|
433 |
+
"epoch": 1.2219959266802445,
|
434 |
+
"grad_norm": 1.0597976659827861,
|
435 |
+
"learning_rate": 3.6046151407029294e-06,
|
436 |
+
"loss": 0.1815,
|
437 |
+
"step": 600
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 1.2423625254582484,
|
441 |
+
"grad_norm": 1.268227483080295,
|
442 |
+
"learning_rate": 3.557664208167006e-06,
|
443 |
+
"loss": 0.182,
|
444 |
+
"step": 610
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 1.2627291242362526,
|
448 |
+
"grad_norm": 1.3674823513528875,
|
449 |
+
"learning_rate": 3.5103060108047724e-06,
|
450 |
+
"loss": 0.181,
|
451 |
+
"step": 620
|
452 |
+
},
|
453 |
+
{
|
454 |
+
"epoch": 1.2830957230142566,
|
455 |
+
"grad_norm": 1.2651346876174,
|
456 |
+
"learning_rate": 3.462564428996006e-06,
|
457 |
+
"loss": 0.1813,
|
458 |
+
"step": 630
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 1.3034623217922607,
|
462 |
+
"grad_norm": 1.1472859968056397,
|
463 |
+
"learning_rate": 3.414463536442161e-06,
|
464 |
+
"loss": 0.1794,
|
465 |
+
"step": 640
|
466 |
+
},
|
467 |
+
{
|
468 |
+
"epoch": 1.3238289205702647,
|
469 |
+
"grad_norm": 1.2499659583847704,
|
470 |
+
"learning_rate": 3.3660275880272014e-06,
|
471 |
+
"loss": 0.1792,
|
472 |
+
"step": 650
|
473 |
+
},
|
474 |
+
{
|
475 |
+
"epoch": 1.3441955193482689,
|
476 |
+
"grad_norm": 1.1361447212288704,
|
477 |
+
"learning_rate": 3.3172810075870724e-06,
|
478 |
+
"loss": 0.1791,
|
479 |
+
"step": 660
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 1.364562118126273,
|
483 |
+
"grad_norm": 1.0924410851905066,
|
484 |
+
"learning_rate": 3.2682483755939735e-06,
|
485 |
+
"loss": 0.1798,
|
486 |
+
"step": 670
|
487 |
+
},
|
488 |
+
{
|
489 |
+
"epoch": 1.384928716904277,
|
490 |
+
"grad_norm": 1.1623338702342625,
|
491 |
+
"learning_rate": 3.2189544167616544e-06,
|
492 |
+
"loss": 0.1796,
|
493 |
+
"step": 680
|
494 |
+
},
|
495 |
+
{
|
496 |
+
"epoch": 1.405295315682281,
|
497 |
+
"grad_norm": 1.3102412387590796,
|
498 |
+
"learning_rate": 3.1694239875779708e-06,
|
499 |
+
"loss": 0.1758,
|
500 |
+
"step": 690
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 1.4256619144602851,
|
504 |
+
"grad_norm": 1.179561417494925,
|
505 |
+
"learning_rate": 3.1196820637709976e-06,
|
506 |
+
"loss": 0.1782,
|
507 |
+
"step": 700
|
508 |
+
},
|
509 |
+
{
|
510 |
+
"epoch": 1.4460285132382893,
|
511 |
+
"grad_norm": 1.1636897041976815,
|
512 |
+
"learning_rate": 3.0697537277150108e-06,
|
513 |
+
"loss": 0.1796,
|
514 |
+
"step": 710
|
515 |
+
},
|
516 |
+
{
|
517 |
+
"epoch": 1.4663951120162932,
|
518 |
+
"grad_norm": 1.0835611859399772,
|
519 |
+
"learning_rate": 3.0196641557826993e-06,
|
520 |
+
"loss": 0.181,
|
521 |
+
"step": 720
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 1.4867617107942974,
|
525 |
+
"grad_norm": 1.1641341412000015,
|
526 |
+
"learning_rate": 2.969438605649973e-06,
|
527 |
+
"loss": 0.1784,
|
528 |
+
"step": 730
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 1.5071283095723014,
|
532 |
+
"grad_norm": 1.4309684703117318,
|
533 |
+
"learning_rate": 2.9191024035597715e-06,
|
534 |
+
"loss": 0.1785,
|
535 |
+
"step": 740
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 1.5274949083503055,
|
539 |
+
"grad_norm": 1.1211318477711534,
|
540 |
+
"learning_rate": 2.868680931551307e-06,
|
541 |
+
"loss": 0.1785,
|
542 |
+
"step": 750
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 1.5478615071283097,
|
546 |
+
"grad_norm": 1.0844733637132042,
|
547 |
+
"learning_rate": 2.818199614661163e-06,
|
548 |
+
"loss": 0.1784,
|
549 |
+
"step": 760
|
550 |
+
},
|
551 |
+
{
|
552 |
+
"epoch": 1.5682281059063137,
|
553 |
+
"grad_norm": 1.0733420166347605,
|
554 |
+
"learning_rate": 2.7676839081027153e-06,
|
555 |
+
"loss": 0.177,
|
556 |
+
"step": 770
|
557 |
+
},
|
558 |
+
{
|
559 |
+
"epoch": 1.5885947046843176,
|
560 |
+
"grad_norm": 1.0406238069755496,
|
561 |
+
"learning_rate": 2.7171592844303377e-06,
|
562 |
+
"loss": 0.177,
|
563 |
+
"step": 780
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 1.6089613034623218,
|
567 |
+
"grad_norm": 1.0173564043856271,
|
568 |
+
"learning_rate": 2.6666512206948574e-06,
|
569 |
+
"loss": 0.1776,
|
570 |
+
"step": 790
|
571 |
+
},
|
572 |
+
{
|
573 |
+
"epoch": 1.629327902240326,
|
574 |
+
"grad_norm": 1.098211071807489,
|
575 |
+
"learning_rate": 2.6161851855967546e-06,
|
576 |
+
"loss": 0.1766,
|
577 |
+
"step": 800
|
578 |
+
},
|
579 |
+
{
|
580 |
+
"epoch": 1.64969450101833,
|
581 |
+
"grad_norm": 1.23874084216799,
|
582 |
+
"learning_rate": 2.565786626643557e-06,
|
583 |
+
"loss": 0.1805,
|
584 |
+
"step": 810
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 1.6700610997963339,
|
588 |
+
"grad_norm": 1.0816973984606886,
|
589 |
+
"learning_rate": 2.5154809573179374e-06,
|
590 |
+
"loss": 0.1797,
|
591 |
+
"step": 820
|
592 |
+
},
|
593 |
+
{
|
594 |
+
"epoch": 1.690427698574338,
|
595 |
+
"grad_norm": 1.0857968496669568,
|
596 |
+
"learning_rate": 2.465293544262953e-06,
|
597 |
+
"loss": 0.1767,
|
598 |
+
"step": 830
|
599 |
+
},
|
600 |
+
{
|
601 |
+
"epoch": 1.7107942973523422,
|
602 |
+
"grad_norm": 1.0614368562971397,
|
603 |
+
"learning_rate": 2.4152496944909185e-06,
|
604 |
+
"loss": 0.1747,
|
605 |
+
"step": 840
|
606 |
+
},
|
607 |
+
{
|
608 |
+
"epoch": 1.7311608961303462,
|
609 |
+
"grad_norm": 1.1079176704743572,
|
610 |
+
"learning_rate": 2.365374642622334e-06,
|
611 |
+
"loss": 0.1749,
|
612 |
+
"step": 850
|
613 |
+
},
|
614 |
+
{
|
615 |
+
"epoch": 1.7515274949083504,
|
616 |
+
"grad_norm": 1.149185145242321,
|
617 |
+
"learning_rate": 2.315693538161327e-06,
|
618 |
+
"loss": 0.1746,
|
619 |
+
"step": 860
|
620 |
+
},
|
621 |
+
{
|
622 |
+
"epoch": 1.7718940936863543,
|
623 |
+
"grad_norm": 1.1364076771423424,
|
624 |
+
"learning_rate": 2.2662314328140077e-06,
|
625 |
+
"loss": 0.1757,
|
626 |
+
"step": 870
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 1.7922606924643585,
|
630 |
+
"grad_norm": 1.0481149270348264,
|
631 |
+
"learning_rate": 2.217013267856143e-06,
|
632 |
+
"loss": 0.175,
|
633 |
+
"step": 880
|
634 |
+
},
|
635 |
+
{
|
636 |
+
"epoch": 1.8126272912423627,
|
637 |
+
"grad_norm": 1.121398301379226,
|
638 |
+
"learning_rate": 2.168063861556515e-06,
|
639 |
+
"loss": 0.1739,
|
640 |
+
"step": 890
|
641 |
+
},
|
642 |
+
{
|
643 |
+
"epoch": 1.8329938900203666,
|
644 |
+
"grad_norm": 1.1627216220108434,
|
645 |
+
"learning_rate": 2.1194078966623004e-06,
|
646 |
+
"loss": 0.1739,
|
647 |
+
"step": 900
|
648 |
+
},
|
649 |
+
{
|
650 |
+
"epoch": 1.8533604887983706,
|
651 |
+
"grad_norm": 1.0423836344485042,
|
652 |
+
"learning_rate": 2.0710699079527936e-06,
|
653 |
+
"loss": 0.1737,
|
654 |
+
"step": 910
|
655 |
+
},
|
656 |
+
{
|
657 |
+
"epoch": 1.8737270875763747,
|
658 |
+
"grad_norm": 1.1321126420156151,
|
659 |
+
"learning_rate": 2.0230742698677407e-06,
|
660 |
+
"loss": 0.1752,
|
661 |
+
"step": 920
|
662 |
+
},
|
663 |
+
{
|
664 |
+
"epoch": 1.894093686354379,
|
665 |
+
"grad_norm": 1.0982860659977216,
|
666 |
+
"learning_rate": 1.9754451842165187e-06,
|
667 |
+
"loss": 0.1732,
|
668 |
+
"step": 930
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 1.9144602851323829,
|
672 |
+
"grad_norm": 1.0607568085093442,
|
673 |
+
"learning_rate": 1.9282066679743756e-06,
|
674 |
+
"loss": 0.1772,
|
675 |
+
"step": 940
|
676 |
+
},
|
677 |
+
{
|
678 |
+
"epoch": 1.9348268839103868,
|
679 |
+
"grad_norm": 1.1195338613324972,
|
680 |
+
"learning_rate": 1.8813825411718608e-06,
|
681 |
+
"loss": 0.1739,
|
682 |
+
"step": 950
|
683 |
+
},
|
684 |
+
{
|
685 |
+
"epoch": 1.955193482688391,
|
686 |
+
"grad_norm": 1.2105165500055401,
|
687 |
+
"learning_rate": 1.834996414883573e-06,
|
688 |
+
"loss": 0.1745,
|
689 |
+
"step": 960
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 1.9755600814663952,
|
693 |
+
"grad_norm": 1.0195074211250463,
|
694 |
+
"learning_rate": 1.789071679322268e-06,
|
695 |
+
"loss": 0.1726,
|
696 |
+
"step": 970
|
697 |
+
},
|
698 |
+
{
|
699 |
+
"epoch": 1.9959266802443993,
|
700 |
+
"grad_norm": 1.1218629665522688,
|
701 |
+
"learning_rate": 1.7436314920443475e-06,
|
702 |
+
"loss": 0.1732,
|
703 |
+
"step": 980
|
704 |
+
},
|
705 |
+
{
|
706 |
+
"epoch": 2.0,
|
707 |
+
"eval_loss": 0.2461756318807602,
|
708 |
+
"eval_runtime": 44.0297,
|
709 |
+
"eval_samples_per_second": 300.275,
|
710 |
+
"eval_steps_per_second": 1.181,
|
711 |
+
"step": 982
|
712 |
+
},
|
713 |
+
{
|
714 |
+
"epoch": 2.016293279022403,
|
715 |
+
"grad_norm": 1.3922840302017268,
|
716 |
+
"learning_rate": 1.6986987662726467e-06,
|
717 |
+
"loss": 0.1202,
|
718 |
+
"step": 990
|
719 |
+
},
|
720 |
+
{
|
721 |
+
"epoch": 2.0366598778004072,
|
722 |
+
"grad_norm": 1.2614971724968627,
|
723 |
+
"learning_rate": 1.6542961593424448e-06,
|
724 |
+
"loss": 0.1048,
|
725 |
+
"step": 1000
|
726 |
+
},
|
727 |
+
{
|
728 |
+
"epoch": 2.0570264765784114,
|
729 |
+
"grad_norm": 1.1248418232642112,
|
730 |
+
"learning_rate": 1.6104460612764963e-06,
|
731 |
+
"loss": 0.1039,
|
732 |
+
"step": 1010
|
733 |
+
},
|
734 |
+
{
|
735 |
+
"epoch": 2.0773930753564156,
|
736 |
+
"grad_norm": 1.1221931057869212,
|
737 |
+
"learning_rate": 1.5671705834948596e-06,
|
738 |
+
"loss": 0.1052,
|
739 |
+
"step": 1020
|
740 |
+
},
|
741 |
+
{
|
742 |
+
"epoch": 2.0977596741344193,
|
743 |
+
"grad_norm": 1.3233889385769144,
|
744 |
+
"learning_rate": 1.5244915476652073e-06,
|
745 |
+
"loss": 0.1045,
|
746 |
+
"step": 1030
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 2.1181262729124235,
|
750 |
+
"grad_norm": 1.219740632707147,
|
751 |
+
"learning_rate": 1.4824304746992488e-06,
|
752 |
+
"loss": 0.1045,
|
753 |
+
"step": 1040
|
754 |
+
},
|
755 |
+
{
|
756 |
+
"epoch": 2.1384928716904277,
|
757 |
+
"grad_norm": 1.1317738274870492,
|
758 |
+
"learning_rate": 1.441008573900804e-06,
|
759 |
+
"loss": 0.1038,
|
760 |
+
"step": 1050
|
761 |
+
},
|
762 |
+
{
|
763 |
+
"epoch": 2.158859470468432,
|
764 |
+
"grad_norm": 1.1214324601378942,
|
765 |
+
"learning_rate": 1.4002467322710097e-06,
|
766 |
+
"loss": 0.1041,
|
767 |
+
"step": 1060
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 2.179226069246436,
|
771 |
+
"grad_norm": 1.113920354990809,
|
772 |
+
"learning_rate": 1.3601655039760452e-06,
|
773 |
+
"loss": 0.1051,
|
774 |
+
"step": 1070
|
775 |
+
},
|
776 |
+
{
|
777 |
+
"epoch": 2.1995926680244398,
|
778 |
+
"grad_norm": 1.1594270616003397,
|
779 |
+
"learning_rate": 1.3207850999826893e-06,
|
780 |
+
"loss": 0.1048,
|
781 |
+
"step": 1080
|
782 |
+
},
|
783 |
+
{
|
784 |
+
"epoch": 2.219959266802444,
|
785 |
+
"grad_norm": 1.026595442656351,
|
786 |
+
"learning_rate": 1.2821253778669374e-06,
|
787 |
+
"loss": 0.104,
|
788 |
+
"step": 1090
|
789 |
+
},
|
790 |
+
{
|
791 |
+
"epoch": 2.240325865580448,
|
792 |
+
"grad_norm": 1.0868748314482166,
|
793 |
+
"learning_rate": 1.2442058318008143e-06,
|
794 |
+
"loss": 0.1045,
|
795 |
+
"step": 1100
|
796 |
+
},
|
797 |
+
{
|
798 |
+
"epoch": 2.2606924643584523,
|
799 |
+
"grad_norm": 1.1163655110568669,
|
800 |
+
"learning_rate": 1.2070455827224361e-06,
|
801 |
+
"loss": 0.1028,
|
802 |
+
"step": 1110
|
803 |
+
},
|
804 |
+
{
|
805 |
+
"epoch": 2.281059063136456,
|
806 |
+
"grad_norm": 1.1265872495516769,
|
807 |
+
"learning_rate": 1.170663368694271e-06,
|
808 |
+
"loss": 0.1049,
|
809 |
+
"step": 1120
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 2.30142566191446,
|
813 |
+
"grad_norm": 1.0787109584417223,
|
814 |
+
"learning_rate": 1.1350775354544707e-06,
|
815 |
+
"loss": 0.104,
|
816 |
+
"step": 1130
|
817 |
+
},
|
818 |
+
{
|
819 |
+
"epoch": 2.3217922606924644,
|
820 |
+
"grad_norm": 1.2647428474671398,
|
821 |
+
"learning_rate": 1.1003060271660259e-06,
|
822 |
+
"loss": 0.104,
|
823 |
+
"step": 1140
|
824 |
+
},
|
825 |
+
{
|
826 |
+
"epoch": 2.3421588594704685,
|
827 |
+
"grad_norm": 1.1925610582272053,
|
828 |
+
"learning_rate": 1.06636637736842e-06,
|
829 |
+
"loss": 0.1042,
|
830 |
+
"step": 1150
|
831 |
+
},
|
832 |
+
{
|
833 |
+
"epoch": 2.3625254582484727,
|
834 |
+
"grad_norm": 1.0856982970751745,
|
835 |
+
"learning_rate": 1.0332757001363402e-06,
|
836 |
+
"loss": 0.1043,
|
837 |
+
"step": 1160
|
838 |
+
},
|
839 |
+
{
|
840 |
+
"epoch": 2.3828920570264764,
|
841 |
+
"grad_norm": 1.1357554208144642,
|
842 |
+
"learning_rate": 1.0010506814498994e-06,
|
843 |
+
"loss": 0.1031,
|
844 |
+
"step": 1170
|
845 |
+
},
|
846 |
+
{
|
847 |
+
"epoch": 2.4032586558044806,
|
848 |
+
"grad_norm": 1.185189241987044,
|
849 |
+
"learning_rate": 9.697075707807282e-07,
|
850 |
+
"loss": 0.1041,
|
851 |
+
"step": 1180
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 2.423625254582485,
|
855 |
+
"grad_norm": 1.134389319376477,
|
856 |
+
"learning_rate": 9.392621728981774e-07,
|
857 |
+
"loss": 0.1045,
|
858 |
+
"step": 1190
|
859 |
+
},
|
860 |
+
{
|
861 |
+
"epoch": 2.443991853360489,
|
862 |
+
"grad_norm": 1.1120088243562505,
|
863 |
+
"learning_rate": 9.097298398997584e-07,
|
864 |
+
"loss": 0.1027,
|
865 |
+
"step": 1200
|
866 |
+
},
|
867 |
+
{
|
868 |
+
"epoch": 2.4643584521384927,
|
869 |
+
"grad_norm": 1.1115040698718408,
|
870 |
+
"learning_rate": 8.811254634698468e-07,
|
871 |
+
"loss": 0.105,
|
872 |
+
"step": 1210
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 2.484725050916497,
|
876 |
+
"grad_norm": 1.0851503339063027,
|
877 |
+
"learning_rate": 8.534634673705461e-07,
|
878 |
+
"loss": 0.1038,
|
879 |
+
"step": 1220
|
880 |
+
},
|
881 |
+
{
|
882 |
+
"epoch": 2.505091649694501,
|
883 |
+
"grad_norm": 1.1183787883183298,
|
884 |
+
"learning_rate": 8.267578001685048e-07,
|
885 |
+
"loss": 0.102,
|
886 |
+
"step": 1230
|
887 |
+
},
|
888 |
+
{
|
889 |
+
"epoch": 2.525458248472505,
|
890 |
+
"grad_norm": 1.1421505006564803,
|
891 |
+
"learning_rate": 8.010219282013471e-07,
|
892 |
+
"loss": 0.1047,
|
893 |
+
"step": 1240
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 2.5458248472505094,
|
897 |
+
"grad_norm": 1.1411521931249191,
|
898 |
+
"learning_rate": 7.76268828787271e-07,
|
899 |
+
"loss": 0.1037,
|
900 |
+
"step": 1250
|
901 |
+
},
|
902 |
+
{
|
903 |
+
"epoch": 2.566191446028513,
|
904 |
+
"grad_norm": 1.0934492001333824,
|
905 |
+
"learning_rate": 7.525109836812336e-07,
|
906 |
+
"loss": 0.1028,
|
907 |
+
"step": 1260
|
908 |
+
},
|
909 |
+
{
|
910 |
+
"epoch": 2.5865580448065173,
|
911 |
+
"grad_norm": 1.098352714811253,
|
912 |
+
"learning_rate": 7.297603727810222e-07,
|
913 |
+
"loss": 0.1034,
|
914 |
+
"step": 1270
|
915 |
+
},
|
916 |
+
{
|
917 |
+
"epoch": 2.6069246435845215,
|
918 |
+
"grad_norm": 1.1012241466696564,
|
919 |
+
"learning_rate": 7.0802846808639e-07,
|
920 |
+
"loss": 0.1039,
|
921 |
+
"step": 1280
|
922 |
+
},
|
923 |
+
{
|
924 |
+
"epoch": 2.627291242362525,
|
925 |
+
"grad_norm": 1.082718927412021,
|
926 |
+
"learning_rate": 6.873262279142977e-07,
|
927 |
+
"loss": 0.1026,
|
928 |
+
"step": 1290
|
929 |
+
},
|
930 |
+
{
|
931 |
+
"epoch": 2.6476578411405294,
|
932 |
+
"grad_norm": 1.0913895786684868,
|
933 |
+
"learning_rate": 6.676640913731799e-07,
|
934 |
+
"loss": 0.1026,
|
935 |
+
"step": 1300
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 2.6680244399185336,
|
939 |
+
"grad_norm": 1.114449983311916,
|
940 |
+
"learning_rate": 6.490519730990235e-07,
|
941 |
+
"loss": 0.1036,
|
942 |
+
"step": 1310
|
943 |
+
},
|
944 |
+
{
|
945 |
+
"epoch": 2.6883910386965377,
|
946 |
+
"grad_norm": 1.0981797148792212,
|
947 |
+
"learning_rate": 6.314992582559093e-07,
|
948 |
+
"loss": 0.1044,
|
949 |
+
"step": 1320
|
950 |
+
},
|
951 |
+
{
|
952 |
+
"epoch": 2.708757637474542,
|
953 |
+
"grad_norm": 1.1031210331032415,
|
954 |
+
"learning_rate": 6.150147978035423e-07,
|
955 |
+
"loss": 0.1028,
|
956 |
+
"step": 1330
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 2.729124236252546,
|
960 |
+
"grad_norm": 1.0231804838279508,
|
961 |
+
"learning_rate": 5.996069040341544e-07,
|
962 |
+
"loss": 0.1007,
|
963 |
+
"step": 1340
|
964 |
+
},
|
965 |
+
{
|
966 |
+
"epoch": 2.74949083503055,
|
967 |
+
"grad_norm": 1.1134674052519227,
|
968 |
+
"learning_rate": 5.852833463810277e-07,
|
969 |
+
"loss": 0.1032,
|
970 |
+
"step": 1350
|
971 |
+
},
|
972 |
+
{
|
973 |
+
"epoch": 2.769857433808554,
|
974 |
+
"grad_norm": 1.0615744074260045,
|
975 |
+
"learning_rate": 5.720513475007569e-07,
|
976 |
+
"loss": 0.1042,
|
977 |
+
"step": 1360
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 2.790224032586558,
|
981 |
+
"grad_norm": 1.0551133440758214,
|
982 |
+
"learning_rate": 5.599175796312243e-07,
|
983 |
+
"loss": 0.1037,
|
984 |
+
"step": 1370
|
985 |
+
},
|
986 |
+
{
|
987 |
+
"epoch": 2.810590631364562,
|
988 |
+
"grad_norm": 1.085953698001666,
|
989 |
+
"learning_rate": 5.488881612271214e-07,
|
990 |
+
"loss": 0.1027,
|
991 |
+
"step": 1380
|
992 |
+
},
|
993 |
+
{
|
994 |
+
"epoch": 2.830957230142566,
|
995 |
+
"grad_norm": 1.069753873055028,
|
996 |
+
"learning_rate": 5.389686538747183e-07,
|
997 |
+
"loss": 0.1036,
|
998 |
+
"step": 1390
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 2.8513238289205702,
|
1002 |
+
"grad_norm": 1.1482168307825666,
|
1003 |
+
"learning_rate": 5.301640594874314e-07,
|
1004 |
+
"loss": 0.1035,
|
1005 |
+
"step": 1400
|
1006 |
+
},
|
1007 |
+
{
|
1008 |
+
"epoch": 2.8716904276985744,
|
1009 |
+
"grad_norm": 1.0724177471486462,
|
1010 |
+
"learning_rate": 5.224788177836086e-07,
|
1011 |
+
"loss": 0.1028,
|
1012 |
+
"step": 1410
|
1013 |
+
},
|
1014 |
+
{
|
1015 |
+
"epoch": 2.8920570264765786,
|
1016 |
+
"grad_norm": 1.0644407890849343,
|
1017 |
+
"learning_rate": 5.159168040478003e-07,
|
1018 |
+
"loss": 0.1036,
|
1019 |
+
"step": 1420
|
1020 |
+
},
|
1021 |
+
{
|
1022 |
+
"epoch": 2.9124236252545828,
|
1023 |
+
"grad_norm": 1.1122712461112416,
|
1024 |
+
"learning_rate": 5.104813271766455e-07,
|
1025 |
+
"loss": 0.1011,
|
1026 |
+
"step": 1430
|
1027 |
+
},
|
1028 |
+
{
|
1029 |
+
"epoch": 2.9327902240325865,
|
1030 |
+
"grad_norm": 1.0864458509689723,
|
1031 |
+
"learning_rate": 5.061751280103621e-07,
|
1032 |
+
"loss": 0.1027,
|
1033 |
+
"step": 1440
|
1034 |
+
},
|
1035 |
+
{
|
1036 |
+
"epoch": 2.9531568228105907,
|
1037 |
+
"grad_norm": 1.0960544633048377,
|
1038 |
+
"learning_rate": 5.030003779506757e-07,
|
1039 |
+
"loss": 0.1016,
|
1040 |
+
"step": 1450
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 2.973523421588595,
|
1044 |
+
"grad_norm": 1.0723267354322656,
|
1045 |
+
"learning_rate": 5.009586778658912e-07,
|
1046 |
+
"loss": 0.1012,
|
1047 |
+
"step": 1460
|
1048 |
+
},
|
1049 |
+
{
|
1050 |
+
"epoch": 2.9938900203665986,
|
1051 |
+
"grad_norm": 1.0958889614425518,
|
1052 |
+
"learning_rate": 5.000510572836534e-07,
|
1053 |
+
"loss": 0.1013,
|
1054 |
+
"step": 1470
|
1055 |
+
},
|
1056 |
+
{
|
1057 |
+
"epoch": 3.0,
|
1058 |
+
"eval_loss": 0.2728930115699768,
|
1059 |
+
"eval_runtime": 45.1101,
|
1060 |
+
"eval_samples_per_second": 293.083,
|
1061 |
+
"eval_steps_per_second": 1.153,
|
1062 |
+
"step": 1473
|
1063 |
+
},
|
1064 |
+
{
|
1065 |
+
"epoch": 3.0,
|
1066 |
+
"step": 1473,
|
1067 |
+
"total_flos": 2467329862533120.0,
|
1068 |
+
"train_loss": 0.19335093406374892,
|
1069 |
+
"train_runtime": 8782.5965,
|
1070 |
+
"train_samples_per_second": 85.801,
|
1071 |
+
"train_steps_per_second": 0.168
|
1072 |
+
}
|
1073 |
+
],
|
1074 |
+
"logging_steps": 10,
|
1075 |
+
"max_steps": 1473,
|
1076 |
+
"num_input_tokens_seen": 0,
|
1077 |
+
"num_train_epochs": 3,
|
1078 |
+
"save_steps": 500,
|
1079 |
+
"stateful_callbacks": {
|
1080 |
+
"TrainerControl": {
|
1081 |
+
"args": {
|
1082 |
+
"should_epoch_stop": false,
|
1083 |
+
"should_evaluate": false,
|
1084 |
+
"should_log": false,
|
1085 |
+
"should_save": true,
|
1086 |
+
"should_training_stop": true
|
1087 |
+
},
|
1088 |
+
"attributes": {}
|
1089 |
+
}
|
1090 |
+
},
|
1091 |
+
"total_flos": 2467329862533120.0,
|
1092 |
+
"train_batch_size": 16,
|
1093 |
+
"trial_name": null,
|
1094 |
+
"trial_params": null
|
1095 |
+
}
|
training_eval_loss.png
ADDED
training_loss.png
ADDED