Training in progress, epoch 1, checkpoint
Browse files
last-checkpoint/adapter_config.json
CHANGED
|
@@ -20,13 +20,13 @@
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
-
"q_proj",
|
| 24 |
-
"o_proj",
|
| 25 |
"up_proj",
|
| 26 |
-
"
|
| 27 |
"k_proj",
|
|
|
|
|
|
|
| 28 |
"down_proj",
|
| 29 |
-
"
|
| 30 |
],
|
| 31 |
"task_type": "CAUSAL_LM",
|
| 32 |
"use_dora": false,
|
|
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
|
|
|
|
|
|
| 23 |
"up_proj",
|
| 24 |
+
"gate_proj",
|
| 25 |
"k_proj",
|
| 26 |
+
"q_proj",
|
| 27 |
+
"o_proj",
|
| 28 |
"down_proj",
|
| 29 |
+
"v_proj"
|
| 30 |
],
|
| 31 |
"task_type": "CAUSAL_LM",
|
| 32 |
"use_dora": false,
|
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 83945296
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dba72936d57e106236c848f5987a7326552c870be18d806f1896dada593a52ac
|
| 3 |
size 83945296
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c005c4b25b09181742e883a4b3e8b822f1c6821cd9b81c0b1f15d5078209824
|
| 3 |
+
size 168037626
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1601a226abaab371fb1a1637d693b97a9298cac19004527736e3a00675bc6cd
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:41bfbb26011cd1f6da678dc46ec4b53dc777067eecadd603a97390cd94b34f6a
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0
|
| 5 |
-
"eval_steps":
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -11,111 +11,2348 @@
|
|
| 11 |
{
|
| 12 |
"epoch": 0.003003003003003003,
|
| 13 |
"grad_norm": NaN,
|
| 14 |
-
"learning_rate": 0.
|
| 15 |
-
"loss": 2.
|
| 16 |
-
"step": 1
|
| 17 |
-
},
|
| 18 |
-
{
|
| 19 |
-
"epoch": 0.003003003003003003,
|
| 20 |
-
"eval_loss": 2.2815349102020264,
|
| 21 |
-
"eval_runtime": 5.7285,
|
| 22 |
-
"eval_samples_per_second": 12.394,
|
| 23 |
-
"eval_steps_per_second": 12.394,
|
| 24 |
"step": 1
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 0.006006006006006006,
|
| 28 |
"grad_norm": NaN,
|
| 29 |
-
"learning_rate": 0.
|
| 30 |
-
"loss": 2.
|
| 31 |
"step": 2
|
| 32 |
},
|
| 33 |
{
|
| 34 |
"epoch": 0.009009009009009009,
|
| 35 |
-
"grad_norm": 8.
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss": 2.
|
| 38 |
-
"step": 3
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.009009009009009009,
|
| 42 |
-
"eval_loss": 2.2815349102020264,
|
| 43 |
-
"eval_runtime": 5.1823,
|
| 44 |
-
"eval_samples_per_second": 13.701,
|
| 45 |
-
"eval_steps_per_second": 13.701,
|
| 46 |
"step": 3
|
| 47 |
},
|
| 48 |
{
|
| 49 |
"epoch": 0.012012012012012012,
|
| 50 |
-
"grad_norm":
|
| 51 |
-
"learning_rate":
|
| 52 |
-
"loss":
|
| 53 |
"step": 4
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 0.015015015015015015,
|
| 57 |
-
"grad_norm": 2.
|
| 58 |
-
"learning_rate":
|
| 59 |
-
"loss": 1.
|
| 60 |
"step": 5
|
| 61 |
},
|
| 62 |
{
|
| 63 |
"epoch": 0.018018018018018018,
|
| 64 |
-
"grad_norm": 4.
|
| 65 |
-
"learning_rate":
|
| 66 |
-
"loss": 2.
|
| 67 |
-
"step": 6
|
| 68 |
-
},
|
| 69 |
-
{
|
| 70 |
-
"epoch": 0.018018018018018018,
|
| 71 |
-
"eval_loss": 2.254021644592285,
|
| 72 |
-
"eval_runtime": 5.358,
|
| 73 |
-
"eval_samples_per_second": 13.251,
|
| 74 |
-
"eval_steps_per_second": 13.251,
|
| 75 |
"step": 6
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.021021021021021023,
|
| 79 |
-
"grad_norm":
|
| 80 |
-
"learning_rate": 0.
|
| 81 |
-
"loss":
|
| 82 |
"step": 7
|
| 83 |
},
|
| 84 |
{
|
| 85 |
"epoch": 0.024024024024024024,
|
| 86 |
-
"grad_norm": 3.
|
| 87 |
-
"learning_rate": 0.
|
| 88 |
-
"loss":
|
| 89 |
"step": 8
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 0.02702702702702703,
|
| 93 |
-
"grad_norm": 3.
|
| 94 |
-
"learning_rate": 0.
|
| 95 |
-
"loss": 1.
|
| 96 |
-
"step": 9
|
| 97 |
-
},
|
| 98 |
-
{
|
| 99 |
-
"epoch": 0.02702702702702703,
|
| 100 |
-
"eval_loss": 2.1258544921875,
|
| 101 |
-
"eval_runtime": 5.0108,
|
| 102 |
-
"eval_samples_per_second": 14.169,
|
| 103 |
-
"eval_steps_per_second": 14.169,
|
| 104 |
"step": 9
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 0.03003003003003003,
|
| 108 |
-
"grad_norm":
|
| 109 |
-
"learning_rate": 0.
|
| 110 |
-
"loss": 2.
|
| 111 |
"step": 10
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 112 |
}
|
| 113 |
],
|
| 114 |
"logging_steps": 1,
|
| 115 |
-
"max_steps":
|
| 116 |
"num_input_tokens_seen": 0,
|
| 117 |
"num_train_epochs": 1,
|
| 118 |
-
"save_steps":
|
| 119 |
"stateful_callbacks": {
|
| 120 |
"TrainerControl": {
|
| 121 |
"args": {
|
|
@@ -128,7 +2365,7 @@
|
|
| 128 |
"attributes": {}
|
| 129 |
}
|
| 130 |
},
|
| 131 |
-
"total_flos":
|
| 132 |
"train_batch_size": 1,
|
| 133 |
"trial_name": null,
|
| 134 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 333,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 11 |
{
|
| 12 |
"epoch": 0.003003003003003003,
|
| 13 |
"grad_norm": NaN,
|
| 14 |
+
"learning_rate": 0.0002,
|
| 15 |
+
"loss": 2.0251,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.006006006006006006,
|
| 20 |
"grad_norm": NaN,
|
| 21 |
+
"learning_rate": 0.0002,
|
| 22 |
+
"loss": 2.8155,
|
| 23 |
"step": 2
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.009009009009009009,
|
| 27 |
+
"grad_norm": 8.254266738891602,
|
| 28 |
+
"learning_rate": 0.0001999955498150411,
|
| 29 |
+
"loss": 2.3407,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
"step": 3
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.012012012012012012,
|
| 34 |
+
"grad_norm": 2.888874053955078,
|
| 35 |
+
"learning_rate": 0.00019998219965624734,
|
| 36 |
+
"loss": 1.982,
|
| 37 |
"step": 4
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.015015015015015015,
|
| 41 |
+
"grad_norm": 2.099008083343506,
|
| 42 |
+
"learning_rate": 0.0001999599507118322,
|
| 43 |
+
"loss": 1.3602,
|
| 44 |
"step": 5
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.018018018018018018,
|
| 48 |
+
"grad_norm": 4.129324913024902,
|
| 49 |
+
"learning_rate": 0.000199928804962034,
|
| 50 |
+
"loss": 2.4564,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
"step": 6
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.021021021021021023,
|
| 55 |
+
"grad_norm": 3.6175684928894043,
|
| 56 |
+
"learning_rate": 0.0001998887651789398,
|
| 57 |
+
"loss": 1.9079,
|
| 58 |
"step": 7
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.024024024024024024,
|
| 62 |
+
"grad_norm": 3.246344804763794,
|
| 63 |
+
"learning_rate": 0.00019983983492623833,
|
| 64 |
+
"loss": 1.6866,
|
| 65 |
"step": 8
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.02702702702702703,
|
| 69 |
+
"grad_norm": 3.2417070865631104,
|
| 70 |
+
"learning_rate": 0.00019978201855890308,
|
| 71 |
+
"loss": 1.4891,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 72 |
"step": 9
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.03003003003003003,
|
| 76 |
+
"grad_norm": 5.3369832038879395,
|
| 77 |
+
"learning_rate": 0.00019971532122280464,
|
| 78 |
+
"loss": 2.0856,
|
| 79 |
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03303303303303303,
|
| 83 |
+
"grad_norm": 3.8723933696746826,
|
| 84 |
+
"learning_rate": 0.00019963974885425266,
|
| 85 |
+
"loss": 1.9191,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.036036036036036036,
|
| 90 |
+
"grad_norm": 5.872828006744385,
|
| 91 |
+
"learning_rate": 0.00019955530817946748,
|
| 92 |
+
"loss": 1.9715,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03903903903903904,
|
| 97 |
+
"grad_norm": 5.758902549743652,
|
| 98 |
+
"learning_rate": 0.0001994620067139815,
|
| 99 |
+
"loss": 1.9539,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.042042042042042045,
|
| 104 |
+
"grad_norm": 4.494828224182129,
|
| 105 |
+
"learning_rate": 0.0001993598527619703,
|
| 106 |
+
"loss": 1.8494,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.04504504504504504,
|
| 111 |
+
"grad_norm": 4.702944755554199,
|
| 112 |
+
"learning_rate": 0.0001992488554155135,
|
| 113 |
+
"loss": 1.7613,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04804804804804805,
|
| 118 |
+
"grad_norm": 4.821094512939453,
|
| 119 |
+
"learning_rate": 0.00019912902455378556,
|
| 120 |
+
"loss": 1.5019,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.05105105105105105,
|
| 125 |
+
"grad_norm": 4.842085838317871,
|
| 126 |
+
"learning_rate": 0.00019900037084217637,
|
| 127 |
+
"loss": 1.014,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.05405405405405406,
|
| 132 |
+
"grad_norm": 4.73835563659668,
|
| 133 |
+
"learning_rate": 0.00019886290573134228,
|
| 134 |
+
"loss": 1.7198,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.057057057057057055,
|
| 139 |
+
"grad_norm": 5.116694927215576,
|
| 140 |
+
"learning_rate": 0.00019871664145618657,
|
| 141 |
+
"loss": 1.029,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.06006006006006006,
|
| 146 |
+
"grad_norm": 3.1407368183135986,
|
| 147 |
+
"learning_rate": 0.00019856159103477086,
|
| 148 |
+
"loss": 1.0887,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.06306306306306306,
|
| 153 |
+
"grad_norm": 4.0600666999816895,
|
| 154 |
+
"learning_rate": 0.00019839776826715614,
|
| 155 |
+
"loss": 1.7268,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.06606606606606606,
|
| 160 |
+
"grad_norm": 5.035831928253174,
|
| 161 |
+
"learning_rate": 0.0001982251877341748,
|
| 162 |
+
"loss": 1.2828,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.06906906906906907,
|
| 167 |
+
"grad_norm": 6.41340446472168,
|
| 168 |
+
"learning_rate": 0.0001980438647961327,
|
| 169 |
+
"loss": 1.6047,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.07207207207207207,
|
| 174 |
+
"grad_norm": 4.085684776306152,
|
| 175 |
+
"learning_rate": 0.00019785381559144196,
|
| 176 |
+
"loss": 1.477,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.07507507507507508,
|
| 181 |
+
"grad_norm": 3.9870212078094482,
|
| 182 |
+
"learning_rate": 0.00019765505703518496,
|
| 183 |
+
"loss": 0.8285,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.07807807807807808,
|
| 188 |
+
"grad_norm": 4.476605415344238,
|
| 189 |
+
"learning_rate": 0.00019744760681760832,
|
| 190 |
+
"loss": 1.3635,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.08108108108108109,
|
| 195 |
+
"grad_norm": 4.659677982330322,
|
| 196 |
+
"learning_rate": 0.00019723148340254892,
|
| 197 |
+
"loss": 1.7153,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.08408408408408409,
|
| 202 |
+
"grad_norm": 4.323594570159912,
|
| 203 |
+
"learning_rate": 0.00019700670602579008,
|
| 204 |
+
"loss": 1.5809,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.08708708708708708,
|
| 209 |
+
"grad_norm": 3.4452884197235107,
|
| 210 |
+
"learning_rate": 0.0001967732946933499,
|
| 211 |
+
"loss": 1.4132,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.09009009009009009,
|
| 216 |
+
"grad_norm": 5.972443103790283,
|
| 217 |
+
"learning_rate": 0.00019653127017970034,
|
| 218 |
+
"loss": 1.6207,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09309309309309309,
|
| 223 |
+
"grad_norm": 2.909363269805908,
|
| 224 |
+
"learning_rate": 0.00019628065402591845,
|
| 225 |
+
"loss": 1.2312,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.0960960960960961,
|
| 230 |
+
"grad_norm": 3.1354382038116455,
|
| 231 |
+
"learning_rate": 0.00019602146853776894,
|
| 232 |
+
"loss": 1.2398,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.0990990990990991,
|
| 237 |
+
"grad_norm": 4.161502361297607,
|
| 238 |
+
"learning_rate": 0.00019575373678371909,
|
| 239 |
+
"loss": 1.5237,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.1021021021021021,
|
| 244 |
+
"grad_norm": 3.4831933975219727,
|
| 245 |
+
"learning_rate": 0.00019547748259288536,
|
| 246 |
+
"loss": 1.3345,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.10510510510510511,
|
| 251 |
+
"grad_norm": 3.050612688064575,
|
| 252 |
+
"learning_rate": 0.00019519273055291266,
|
| 253 |
+
"loss": 1.233,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.10810810810810811,
|
| 258 |
+
"grad_norm": 3.520677089691162,
|
| 259 |
+
"learning_rate": 0.0001948995060077859,
|
| 260 |
+
"loss": 1.3311,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.1111111111111111,
|
| 265 |
+
"grad_norm": 3.586829423904419,
|
| 266 |
+
"learning_rate": 0.00019459783505557424,
|
| 267 |
+
"loss": 1.212,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.11411411411411411,
|
| 272 |
+
"grad_norm": 2.9168314933776855,
|
| 273 |
+
"learning_rate": 0.00019428774454610843,
|
| 274 |
+
"loss": 1.084,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.11711711711711711,
|
| 279 |
+
"grad_norm": 3.1117663383483887,
|
| 280 |
+
"learning_rate": 0.00019396926207859084,
|
| 281 |
+
"loss": 0.978,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.12012012012012012,
|
| 286 |
+
"grad_norm": 3.9365386962890625,
|
| 287 |
+
"learning_rate": 0.00019364241599913924,
|
| 288 |
+
"loss": 0.91,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.12312312312312312,
|
| 293 |
+
"grad_norm": 5.414772987365723,
|
| 294 |
+
"learning_rate": 0.00019330723539826375,
|
| 295 |
+
"loss": 1.5547,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.12612612612612611,
|
| 300 |
+
"grad_norm": 3.866046905517578,
|
| 301 |
+
"learning_rate": 0.00019296375010827773,
|
| 302 |
+
"loss": 1.3311,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.12912912912912913,
|
| 307 |
+
"grad_norm": 4.769541263580322,
|
| 308 |
+
"learning_rate": 0.0001926119907006426,
|
| 309 |
+
"loss": 0.8167,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.13213213213213212,
|
| 314 |
+
"grad_norm": 3.6581079959869385,
|
| 315 |
+
"learning_rate": 0.0001922519884832469,
|
| 316 |
+
"loss": 1.2906,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.13513513513513514,
|
| 321 |
+
"grad_norm": 4.367158889770508,
|
| 322 |
+
"learning_rate": 0.00019188377549761963,
|
| 323 |
+
"loss": 1.2016,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.13813813813813813,
|
| 328 |
+
"grad_norm": 3.591726541519165,
|
| 329 |
+
"learning_rate": 0.0001915073845160786,
|
| 330 |
+
"loss": 1.4798,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.14114114114114115,
|
| 335 |
+
"grad_norm": 4.844123840332031,
|
| 336 |
+
"learning_rate": 0.0001911228490388136,
|
| 337 |
+
"loss": 1.4799,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.14414414414414414,
|
| 342 |
+
"grad_norm": 5.353418350219727,
|
| 343 |
+
"learning_rate": 0.00019073020329090444,
|
| 344 |
+
"loss": 1.2234,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.14714714714714713,
|
| 349 |
+
"grad_norm": 3.3358774185180664,
|
| 350 |
+
"learning_rate": 0.00019032948221927524,
|
| 351 |
+
"loss": 1.0556,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.15015015015015015,
|
| 356 |
+
"grad_norm": 3.674219846725464,
|
| 357 |
+
"learning_rate": 0.00018992072148958368,
|
| 358 |
+
"loss": 1.1757,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.15315315315315314,
|
| 363 |
+
"grad_norm": 4.523245334625244,
|
| 364 |
+
"learning_rate": 0.00018950395748304678,
|
| 365 |
+
"loss": 1.2357,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.15615615615615616,
|
| 370 |
+
"grad_norm": 4.535871982574463,
|
| 371 |
+
"learning_rate": 0.00018907922729320285,
|
| 372 |
+
"loss": 1.5018,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.15915915915915915,
|
| 377 |
+
"grad_norm": 3.2581071853637695,
|
| 378 |
+
"learning_rate": 0.00018864656872260985,
|
| 379 |
+
"loss": 1.1452,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.16216216216216217,
|
| 384 |
+
"grad_norm": 3.132718324661255,
|
| 385 |
+
"learning_rate": 0.00018820602027948114,
|
| 386 |
+
"loss": 1.1584,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.16516516516516516,
|
| 391 |
+
"grad_norm": 4.070220470428467,
|
| 392 |
+
"learning_rate": 0.00018775762117425777,
|
| 393 |
+
"loss": 1.1046,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.16816816816816818,
|
| 398 |
+
"grad_norm": 3.8557350635528564,
|
| 399 |
+
"learning_rate": 0.00018730141131611882,
|
| 400 |
+
"loss": 0.9695,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.17117117117117117,
|
| 405 |
+
"grad_norm": 3.3533196449279785,
|
| 406 |
+
"learning_rate": 0.00018683743130942928,
|
| 407 |
+
"loss": 1.1877,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.17417417417417416,
|
| 412 |
+
"grad_norm": 3.0984268188476562,
|
| 413 |
+
"learning_rate": 0.00018636572245012606,
|
| 414 |
+
"loss": 1.3991,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.17717717717717718,
|
| 419 |
+
"grad_norm": 10.649101257324219,
|
| 420 |
+
"learning_rate": 0.00018588632672204264,
|
| 421 |
+
"loss": 1.1049,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.18018018018018017,
|
| 426 |
+
"grad_norm": 4.183538436889648,
|
| 427 |
+
"learning_rate": 0.0001853992867931721,
|
| 428 |
+
"loss": 1.4726,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.1831831831831832,
|
| 433 |
+
"grad_norm": 3.4067232608795166,
|
| 434 |
+
"learning_rate": 0.0001849046460118698,
|
| 435 |
+
"loss": 1.3283,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.18618618618618618,
|
| 440 |
+
"grad_norm": 3.2232909202575684,
|
| 441 |
+
"learning_rate": 0.00018440244840299506,
|
| 442 |
+
"loss": 1.0828,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.1891891891891892,
|
| 447 |
+
"grad_norm": 3.587733030319214,
|
| 448 |
+
"learning_rate": 0.00018389273866399275,
|
| 449 |
+
"loss": 1.4606,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.1921921921921922,
|
| 454 |
+
"grad_norm": 3.715364456176758,
|
| 455 |
+
"learning_rate": 0.00018337556216091517,
|
| 456 |
+
"loss": 1.3308,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.19519519519519518,
|
| 461 |
+
"grad_norm": 3.0346460342407227,
|
| 462 |
+
"learning_rate": 0.00018285096492438424,
|
| 463 |
+
"loss": 1.1785,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.1981981981981982,
|
| 468 |
+
"grad_norm": 2.9115614891052246,
|
| 469 |
+
"learning_rate": 0.00018231899364549455,
|
| 470 |
+
"loss": 0.8434,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2012012012012012,
|
| 475 |
+
"grad_norm": 3.1807260513305664,
|
| 476 |
+
"learning_rate": 0.0001817796956716578,
|
| 477 |
+
"loss": 1.0018,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2042042042042042,
|
| 482 |
+
"grad_norm": 3.0817840099334717,
|
| 483 |
+
"learning_rate": 0.0001812331190023886,
|
| 484 |
+
"loss": 1.0185,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2072072072072072,
|
| 489 |
+
"grad_norm": 3.2719409465789795,
|
| 490 |
+
"learning_rate": 0.00018067931228503246,
|
| 491 |
+
"loss": 1.4434,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.21021021021021022,
|
| 496 |
+
"grad_norm": 4.187236309051514,
|
| 497 |
+
"learning_rate": 0.00018011832481043576,
|
| 498 |
+
"loss": 1.0969,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2132132132132132,
|
| 503 |
+
"grad_norm": 3.2186405658721924,
|
| 504 |
+
"learning_rate": 0.000179550206508559,
|
| 505 |
+
"loss": 1.2105,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.21621621621621623,
|
| 510 |
+
"grad_norm": 3.406670570373535,
|
| 511 |
+
"learning_rate": 0.0001789750079440326,
|
| 512 |
+
"loss": 1.3271,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.21921921921921922,
|
| 517 |
+
"grad_norm": 3.0242960453033447,
|
| 518 |
+
"learning_rate": 0.00017839278031165658,
|
| 519 |
+
"loss": 1.1269,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2222222222222222,
|
| 524 |
+
"grad_norm": 4.0513691902160645,
|
| 525 |
+
"learning_rate": 0.00017780357543184397,
|
| 526 |
+
"loss": 1.1778,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.22522522522522523,
|
| 531 |
+
"grad_norm": 4.116867542266846,
|
| 532 |
+
"learning_rate": 0.00017720744574600863,
|
| 533 |
+
"loss": 1.2357,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.22822822822822822,
|
| 538 |
+
"grad_norm": 3.957306146621704,
|
| 539 |
+
"learning_rate": 0.0001766044443118978,
|
| 540 |
+
"loss": 1.4235,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.23123123123123124,
|
| 545 |
+
"grad_norm": 7.28619384765625,
|
| 546 |
+
"learning_rate": 0.00017599462479886974,
|
| 547 |
+
"loss": 1.102,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.23423423423423423,
|
| 552 |
+
"grad_norm": 4.001276969909668,
|
| 553 |
+
"learning_rate": 0.00017537804148311695,
|
| 554 |
+
"loss": 1.1948,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.23723723723723725,
|
| 559 |
+
"grad_norm": 2.7599635124206543,
|
| 560 |
+
"learning_rate": 0.00017475474924283536,
|
| 561 |
+
"loss": 1.0386,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.24024024024024024,
|
| 566 |
+
"grad_norm": 3.36354923248291,
|
| 567 |
+
"learning_rate": 0.00017412480355334005,
|
| 568 |
+
"loss": 1.6077,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.24324324324324326,
|
| 573 |
+
"grad_norm": NaN,
|
| 574 |
+
"learning_rate": 0.00017412480355334005,
|
| 575 |
+
"loss": 1.4291,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.24624624624624625,
|
| 580 |
+
"grad_norm": 4.241931915283203,
|
| 581 |
+
"learning_rate": 0.0001734882604821276,
|
| 582 |
+
"loss": 1.37,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.24924924924924924,
|
| 587 |
+
"grad_norm": 3.40556001663208,
|
| 588 |
+
"learning_rate": 0.0001728451766838861,
|
| 589 |
+
"loss": 1.4526,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.25225225225225223,
|
| 594 |
+
"grad_norm": 3.1627070903778076,
|
| 595 |
+
"learning_rate": 0.00017219560939545246,
|
| 596 |
+
"loss": 1.1918,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.2552552552552553,
|
| 601 |
+
"grad_norm": 2.915019989013672,
|
| 602 |
+
"learning_rate": 0.0001715396164307182,
|
| 603 |
+
"loss": 1.1523,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.25825825825825827,
|
| 608 |
+
"grad_norm": 3.027531862258911,
|
| 609 |
+
"learning_rate": 0.00017087725617548385,
|
| 610 |
+
"loss": 1.097,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.26126126126126126,
|
| 615 |
+
"grad_norm": 3.198509931564331,
|
| 616 |
+
"learning_rate": 0.00017020858758226229,
|
| 617 |
+
"loss": 1.1172,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.26426426426426425,
|
| 622 |
+
"grad_norm": 2.9282870292663574,
|
| 623 |
+
"learning_rate": 0.00016953367016503182,
|
| 624 |
+
"loss": 0.9913,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.2672672672672673,
|
| 629 |
+
"grad_norm": 2.8152143955230713,
|
| 630 |
+
"learning_rate": 0.00016885256399393924,
|
| 631 |
+
"loss": 0.9582,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.2702702702702703,
|
| 636 |
+
"grad_norm": 5.513664245605469,
|
| 637 |
+
"learning_rate": 0.00016816532968995328,
|
| 638 |
+
"loss": 1.5983,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.2732732732732733,
|
| 643 |
+
"grad_norm": 3.0489561557769775,
|
| 644 |
+
"learning_rate": 0.00016747202841946928,
|
| 645 |
+
"loss": 1.3503,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.27627627627627627,
|
| 650 |
+
"grad_norm": 3.1149275302886963,
|
| 651 |
+
"learning_rate": 0.00016677272188886483,
|
| 652 |
+
"loss": 1.0439,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.27927927927927926,
|
| 657 |
+
"grad_norm": 5.150669097900391,
|
| 658 |
+
"learning_rate": 0.00016606747233900815,
|
| 659 |
+
"loss": 1.5164,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.2822822822822823,
|
| 664 |
+
"grad_norm": 3.6513853073120117,
|
| 665 |
+
"learning_rate": 0.00016535634253971794,
|
| 666 |
+
"loss": 0.8632,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.2852852852852853,
|
| 671 |
+
"grad_norm": 2.818716287612915,
|
| 672 |
+
"learning_rate": 0.00016463939578417692,
|
| 673 |
+
"loss": 0.9685,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.2882882882882883,
|
| 678 |
+
"grad_norm": 4.800100803375244,
|
| 679 |
+
"learning_rate": 0.0001639166958832985,
|
| 680 |
+
"loss": 1.6414,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.2912912912912913,
|
| 685 |
+
"grad_norm": 5.5120649337768555,
|
| 686 |
+
"learning_rate": 0.00016318830716004722,
|
| 687 |
+
"loss": 2.0318,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.29429429429429427,
|
| 692 |
+
"grad_norm": 3.650167226791382,
|
| 693 |
+
"learning_rate": 0.0001624542944437139,
|
| 694 |
+
"loss": 1.3504,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.2972972972972973,
|
| 699 |
+
"grad_norm": 3.262024402618408,
|
| 700 |
+
"learning_rate": 0.00016171472306414554,
|
| 701 |
+
"loss": 0.9678,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3003003003003003,
|
| 706 |
+
"grad_norm": 3.3580527305603027,
|
| 707 |
+
"learning_rate": 0.0001609696588459307,
|
| 708 |
+
"loss": 1.2071,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3033033033033033,
|
| 713 |
+
"grad_norm": 2.174422025680542,
|
| 714 |
+
"learning_rate": 0.00016021916810254097,
|
| 715 |
+
"loss": 0.8678,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.3063063063063063,
|
| 720 |
+
"grad_norm": 3.338526487350464,
|
| 721 |
+
"learning_rate": 0.00015946331763042867,
|
| 722 |
+
"loss": 1.1964,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.30930930930930933,
|
| 727 |
+
"grad_norm": 2.4254307746887207,
|
| 728 |
+
"learning_rate": 0.00015870217470308188,
|
| 729 |
+
"loss": 1.1689,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3123123123123123,
|
| 734 |
+
"grad_norm": 3.5244359970092773,
|
| 735 |
+
"learning_rate": 0.0001579358070650367,
|
| 736 |
+
"loss": 1.5078,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.3153153153153153,
|
| 741 |
+
"grad_norm": 3.7619662284851074,
|
| 742 |
+
"learning_rate": 0.00015716428292584787,
|
| 743 |
+
"loss": 1.2496,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3183183183183183,
|
| 748 |
+
"grad_norm": 4.3766303062438965,
|
| 749 |
+
"learning_rate": 0.0001563876709540178,
|
| 750 |
+
"loss": 1.673,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3213213213213213,
|
| 755 |
+
"grad_norm": 1.925053596496582,
|
| 756 |
+
"learning_rate": 0.00015560604027088477,
|
| 757 |
+
"loss": 0.7445,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.32432432432432434,
|
| 762 |
+
"grad_norm": 2.5130960941314697,
|
| 763 |
+
"learning_rate": 0.00015481946044447099,
|
| 764 |
+
"loss": 0.8784,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.32732732732732733,
|
| 769 |
+
"grad_norm": 3.086637020111084,
|
| 770 |
+
"learning_rate": 0.00015402800148329071,
|
| 771 |
+
"loss": 1.1608,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.3303303303303303,
|
| 776 |
+
"grad_norm": 2.877929925918579,
|
| 777 |
+
"learning_rate": 0.0001532317338301192,
|
| 778 |
+
"loss": 0.9467,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3333333333333333,
|
| 783 |
+
"grad_norm": 2.76326847076416,
|
| 784 |
+
"learning_rate": 0.00015243072835572318,
|
| 785 |
+
"loss": 1.0425,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.33633633633633636,
|
| 790 |
+
"grad_norm": 3.9744467735290527,
|
| 791 |
+
"learning_rate": 0.00015162505635255287,
|
| 792 |
+
"loss": 1.3432,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.33933933933933935,
|
| 797 |
+
"grad_norm": 5.023432731628418,
|
| 798 |
+
"learning_rate": 0.00015081478952839693,
|
| 799 |
+
"loss": 1.6413,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.34234234234234234,
|
| 804 |
+
"grad_norm": 3.5280025005340576,
|
| 805 |
+
"learning_rate": 0.00015000000000000001,
|
| 806 |
+
"loss": 1.0987,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.34534534534534533,
|
| 811 |
+
"grad_norm": 2.802316665649414,
|
| 812 |
+
"learning_rate": 0.0001491807602866442,
|
| 813 |
+
"loss": 0.7508,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.3483483483483483,
|
| 818 |
+
"grad_norm": 3.0784826278686523,
|
| 819 |
+
"learning_rate": 0.00014835714330369446,
|
| 820 |
+
"loss": 1.199,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.35135135135135137,
|
| 825 |
+
"grad_norm": 4.75717306137085,
|
| 826 |
+
"learning_rate": 0.000147529222356109,
|
| 827 |
+
"loss": 1.3701,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.35435435435435436,
|
| 832 |
+
"grad_norm": 2.9851152896881104,
|
| 833 |
+
"learning_rate": 0.00014669707113191483,
|
| 834 |
+
"loss": 1.037,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.35735735735735735,
|
| 839 |
+
"grad_norm": 4.035271644592285,
|
| 840 |
+
"learning_rate": 0.00014586076369564908,
|
| 841 |
+
"loss": 1.0533,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.36036036036036034,
|
| 846 |
+
"grad_norm": 9.006219863891602,
|
| 847 |
+
"learning_rate": 0.00014502037448176734,
|
| 848 |
+
"loss": 1.0655,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3633633633633634,
|
| 853 |
+
"grad_norm": 3.5663275718688965,
|
| 854 |
+
"learning_rate": 0.00014417597828801832,
|
| 855 |
+
"loss": 1.568,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.3663663663663664,
|
| 860 |
+
"grad_norm": 2.996906280517578,
|
| 861 |
+
"learning_rate": 0.00014332765026878687,
|
| 862 |
+
"loss": 1.0472,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.36936936936936937,
|
| 867 |
+
"grad_norm": 2.879059076309204,
|
| 868 |
+
"learning_rate": 0.0001424754659284048,
|
| 869 |
+
"loss": 1.214,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.37237237237237236,
|
| 874 |
+
"grad_norm": 3.065863847732544,
|
| 875 |
+
"learning_rate": 0.00014161950111443077,
|
| 876 |
+
"loss": 0.8353,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.37537537537537535,
|
| 881 |
+
"grad_norm": 4.549298286437988,
|
| 882 |
+
"learning_rate": 0.00014075983201089964,
|
| 883 |
+
"loss": 0.9154,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.3783783783783784,
|
| 888 |
+
"grad_norm": 4.853454113006592,
|
| 889 |
+
"learning_rate": 0.00013989653513154165,
|
| 890 |
+
"loss": 1.5181,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.3813813813813814,
|
| 895 |
+
"grad_norm": 2.2490577697753906,
|
| 896 |
+
"learning_rate": 0.00013902968731297255,
|
| 897 |
+
"loss": 0.9221,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3843843843843844,
|
| 902 |
+
"grad_norm": 4.746774673461914,
|
| 903 |
+
"learning_rate": 0.00013815936570785487,
|
| 904 |
+
"loss": 0.7953,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.38738738738738737,
|
| 909 |
+
"grad_norm": 3.5353190898895264,
|
| 910 |
+
"learning_rate": 0.00013728564777803088,
|
| 911 |
+
"loss": 0.9101,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.39039039039039036,
|
| 916 |
+
"grad_norm": 3.052245855331421,
|
| 917 |
+
"learning_rate": 0.0001364086112876284,
|
| 918 |
+
"loss": 1.3159,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.3933933933933934,
|
| 923 |
+
"grad_norm": 4.477910041809082,
|
| 924 |
+
"learning_rate": 0.00013552833429613938,
|
| 925 |
+
"loss": 1.2709,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.3963963963963964,
|
| 930 |
+
"grad_norm": 7.228553295135498,
|
| 931 |
+
"learning_rate": 0.00013464489515147238,
|
| 932 |
+
"loss": 1.4622,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.3993993993993994,
|
| 937 |
+
"grad_norm": 2.7557127475738525,
|
| 938 |
+
"learning_rate": 0.00013375837248297926,
|
| 939 |
+
"loss": 1.4031,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.4024024024024024,
|
| 944 |
+
"grad_norm": 5.027473449707031,
|
| 945 |
+
"learning_rate": 0.0001328688451944569,
|
| 946 |
+
"loss": 1.2564,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.40540540540540543,
|
| 951 |
+
"grad_norm": 2.9252965450286865,
|
| 952 |
+
"learning_rate": 0.00013197639245712454,
|
| 953 |
+
"loss": 1.0321,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.4084084084084084,
|
| 958 |
+
"grad_norm": 2.841583728790283,
|
| 959 |
+
"learning_rate": 0.00013108109370257712,
|
| 960 |
+
"loss": 0.9534,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.4114114114114114,
|
| 965 |
+
"grad_norm": 4.3325114250183105,
|
| 966 |
+
"learning_rate": 0.0001301830286157157,
|
| 967 |
+
"loss": 1.4278,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.4144144144144144,
|
| 972 |
+
"grad_norm": 3.721039056777954,
|
| 973 |
+
"learning_rate": 0.00012928227712765504,
|
| 974 |
+
"loss": 1.4167,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.4174174174174174,
|
| 979 |
+
"grad_norm": 4.287530422210693,
|
| 980 |
+
"learning_rate": 0.00012837891940860972,
|
| 981 |
+
"loss": 1.4874,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.42042042042042044,
|
| 986 |
+
"grad_norm": 3.366671085357666,
|
| 987 |
+
"learning_rate": 0.0001274730358607583,
|
| 988 |
+
"loss": 1.0406,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.42342342342342343,
|
| 993 |
+
"grad_norm": 4.121444225311279,
|
| 994 |
+
"learning_rate": 0.00012656470711108764,
|
| 995 |
+
"loss": 1.6264,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.4264264264264264,
|
| 1000 |
+
"grad_norm": 3.2435896396636963,
|
| 1001 |
+
"learning_rate": 0.00012565401400421651,
|
| 1002 |
+
"loss": 1.1203,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.4294294294294294,
|
| 1007 |
+
"grad_norm": 2.8955342769622803,
|
| 1008 |
+
"learning_rate": 0.00012474103759520027,
|
| 1009 |
+
"loss": 1.2582,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.43243243243243246,
|
| 1014 |
+
"grad_norm": 4.04188346862793,
|
| 1015 |
+
"learning_rate": 0.0001238258591423165,
|
| 1016 |
+
"loss": 1.0391,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.43543543543543545,
|
| 1021 |
+
"grad_norm": 4.958117961883545,
|
| 1022 |
+
"learning_rate": 0.000122908560099833,
|
| 1023 |
+
"loss": 1.4298,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.43843843843843844,
|
| 1028 |
+
"grad_norm": 1.997272253036499,
|
| 1029 |
+
"learning_rate": 0.00012198922211075778,
|
| 1030 |
+
"loss": 0.9793,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.44144144144144143,
|
| 1035 |
+
"grad_norm": 2.4838814735412598,
|
| 1036 |
+
"learning_rate": 0.00012106792699957263,
|
| 1037 |
+
"loss": 1.0245,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.4444444444444444,
|
| 1042 |
+
"grad_norm": 3.6618947982788086,
|
| 1043 |
+
"learning_rate": 0.00012014475676495052,
|
| 1044 |
+
"loss": 1.2034,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.44744744744744747,
|
| 1049 |
+
"grad_norm": 2.2039713859558105,
|
| 1050 |
+
"learning_rate": 0.0001192197935724573,
|
| 1051 |
+
"loss": 0.9593,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.45045045045045046,
|
| 1056 |
+
"grad_norm": 3.798941135406494,
|
| 1057 |
+
"learning_rate": 0.00011829311974723867,
|
| 1058 |
+
"loss": 1.6691,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.45345345345345345,
|
| 1063 |
+
"grad_norm": 1.9355311393737793,
|
| 1064 |
+
"learning_rate": 0.00011736481776669306,
|
| 1065 |
+
"loss": 0.6611,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.45645645645645644,
|
| 1070 |
+
"grad_norm": 2.3967108726501465,
|
| 1071 |
+
"learning_rate": 0.00011643497025313061,
|
| 1072 |
+
"loss": 1.0664,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.4594594594594595,
|
| 1077 |
+
"grad_norm": 4.215780735015869,
|
| 1078 |
+
"learning_rate": 0.00011550365996641979,
|
| 1079 |
+
"loss": 0.9672,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.4624624624624625,
|
| 1084 |
+
"grad_norm": 3.002946615219116,
|
| 1085 |
+
"learning_rate": 0.00011457096979662114,
|
| 1086 |
+
"loss": 1.6195,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.46546546546546547,
|
| 1091 |
+
"grad_norm": 2.1367151737213135,
|
| 1092 |
+
"learning_rate": 0.00011363698275661001,
|
| 1093 |
+
"loss": 0.7318,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.46846846846846846,
|
| 1098 |
+
"grad_norm": 2.3746261596679688,
|
| 1099 |
+
"learning_rate": 0.00011270178197468789,
|
| 1100 |
+
"loss": 0.7527,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.47147147147147145,
|
| 1105 |
+
"grad_norm": 4.308503150939941,
|
| 1106 |
+
"learning_rate": 0.00011176545068718385,
|
| 1107 |
+
"loss": 1.6989,
|
| 1108 |
+
"step": 157
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.4744744744744745,
|
| 1112 |
+
"grad_norm": 2.32792067527771,
|
| 1113 |
+
"learning_rate": 0.0001108280722310462,
|
| 1114 |
+
"loss": 1.262,
|
| 1115 |
+
"step": 158
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.4774774774774775,
|
| 1119 |
+
"grad_norm": 4.225987911224365,
|
| 1120 |
+
"learning_rate": 0.00010988973003642499,
|
| 1121 |
+
"loss": 1.4396,
|
| 1122 |
+
"step": 159
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4804804804804805,
|
| 1126 |
+
"grad_norm": 3.8827497959136963,
|
| 1127 |
+
"learning_rate": 0.00010895050761924668,
|
| 1128 |
+
"loss": 1.5353,
|
| 1129 |
+
"step": 160
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.48348348348348347,
|
| 1133 |
+
"grad_norm": 3.854318380355835,
|
| 1134 |
+
"learning_rate": 0.00010801048857378071,
|
| 1135 |
+
"loss": 1.4654,
|
| 1136 |
+
"step": 161
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.4864864864864865,
|
| 1140 |
+
"grad_norm": 2.3658535480499268,
|
| 1141 |
+
"learning_rate": 0.00010706975656519946,
|
| 1142 |
+
"loss": 1.0751,
|
| 1143 |
+
"step": 162
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.4894894894894895,
|
| 1147 |
+
"grad_norm": 6.701672554016113,
|
| 1148 |
+
"learning_rate": 0.00010612839532213164,
|
| 1149 |
+
"loss": 1.6829,
|
| 1150 |
+
"step": 163
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.4924924924924925,
|
| 1154 |
+
"grad_norm": 3.207954168319702,
|
| 1155 |
+
"learning_rate": 0.00010518648862921012,
|
| 1156 |
+
"loss": 1.4072,
|
| 1157 |
+
"step": 164
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.4954954954954955,
|
| 1161 |
+
"grad_norm": 2.800644636154175,
|
| 1162 |
+
"learning_rate": 0.00010424412031961484,
|
| 1163 |
+
"loss": 1.0728,
|
| 1164 |
+
"step": 165
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.4984984984984985,
|
| 1168 |
+
"grad_norm": 4.086751461029053,
|
| 1169 |
+
"learning_rate": 0.00010330137426761135,
|
| 1170 |
+
"loss": 1.1628,
|
| 1171 |
+
"step": 166
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5015015015015015,
|
| 1175 |
+
"grad_norm": 2.659355640411377,
|
| 1176 |
+
"learning_rate": 0.00010235833438108571,
|
| 1177 |
+
"loss": 1.1724,
|
| 1178 |
+
"step": 167
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.5045045045045045,
|
| 1182 |
+
"grad_norm": 3.152151107788086,
|
| 1183 |
+
"learning_rate": 0.00010141508459407623,
|
| 1184 |
+
"loss": 1.2078,
|
| 1185 |
+
"step": 168
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5075075075075075,
|
| 1189 |
+
"grad_norm": 3.8074827194213867,
|
| 1190 |
+
"learning_rate": 0.00010047170885930324,
|
| 1191 |
+
"loss": 1.1832,
|
| 1192 |
+
"step": 169
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5105105105105106,
|
| 1196 |
+
"grad_norm": 3.1562180519104004,
|
| 1197 |
+
"learning_rate": 9.95282911406968e-05,
|
| 1198 |
+
"loss": 1.1005,
|
| 1199 |
+
"step": 170
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5135135135135135,
|
| 1203 |
+
"grad_norm": 2.793989419937134,
|
| 1204 |
+
"learning_rate": 9.858491540592382e-05,
|
| 1205 |
+
"loss": 1.0381,
|
| 1206 |
+
"step": 171
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5165165165165165,
|
| 1210 |
+
"grad_norm": 3.589442491531372,
|
| 1211 |
+
"learning_rate": 9.764166561891432e-05,
|
| 1212 |
+
"loss": 0.9335,
|
| 1213 |
+
"step": 172
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5195195195195195,
|
| 1217 |
+
"grad_norm": 4.19896125793457,
|
| 1218 |
+
"learning_rate": 9.669862573238863e-05,
|
| 1219 |
+
"loss": 1.1411,
|
| 1220 |
+
"step": 173
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5225225225225225,
|
| 1224 |
+
"grad_norm": 3.931088924407959,
|
| 1225 |
+
"learning_rate": 9.57558796803852e-05,
|
| 1226 |
+
"loss": 1.0498,
|
| 1227 |
+
"step": 174
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5255255255255256,
|
| 1231 |
+
"grad_norm": 2.746661424636841,
|
| 1232 |
+
"learning_rate": 9.48135113707899e-05,
|
| 1233 |
+
"loss": 1.0747,
|
| 1234 |
+
"step": 175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5285285285285285,
|
| 1238 |
+
"grad_norm": 2.2940986156463623,
|
| 1239 |
+
"learning_rate": 9.38716046778684e-05,
|
| 1240 |
+
"loss": 1.0899,
|
| 1241 |
+
"step": 176
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.5315315315315315,
|
| 1245 |
+
"grad_norm": 2.696507692337036,
|
| 1246 |
+
"learning_rate": 9.293024343480055e-05,
|
| 1247 |
+
"loss": 0.7636,
|
| 1248 |
+
"step": 177
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5345345345345346,
|
| 1252 |
+
"grad_norm": 2.972304344177246,
|
| 1253 |
+
"learning_rate": 9.198951142621929e-05,
|
| 1254 |
+
"loss": 1.349,
|
| 1255 |
+
"step": 178
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5375375375375375,
|
| 1259 |
+
"grad_norm": 2.4269001483917236,
|
| 1260 |
+
"learning_rate": 9.104949238075336e-05,
|
| 1261 |
+
"loss": 1.1242,
|
| 1262 |
+
"step": 179
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.5405405405405406,
|
| 1266 |
+
"grad_norm": 3.105961799621582,
|
| 1267 |
+
"learning_rate": 9.011026996357503e-05,
|
| 1268 |
+
"loss": 1.0303,
|
| 1269 |
+
"step": 180
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5435435435435435,
|
| 1273 |
+
"grad_norm": 2.861002206802368,
|
| 1274 |
+
"learning_rate": 8.917192776895382e-05,
|
| 1275 |
+
"loss": 1.2752,
|
| 1276 |
+
"step": 181
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5465465465465466,
|
| 1280 |
+
"grad_norm": 3.16262149810791,
|
| 1281 |
+
"learning_rate": 8.823454931281616e-05,
|
| 1282 |
+
"loss": 0.9932,
|
| 1283 |
+
"step": 182
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.5495495495495496,
|
| 1287 |
+
"grad_norm": 2.2218563556671143,
|
| 1288 |
+
"learning_rate": 8.729821802531212e-05,
|
| 1289 |
+
"loss": 0.9025,
|
| 1290 |
+
"step": 183
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.5525525525525525,
|
| 1294 |
+
"grad_norm": 3.3385350704193115,
|
| 1295 |
+
"learning_rate": 8.636301724339004e-05,
|
| 1296 |
+
"loss": 1.1192,
|
| 1297 |
+
"step": 184
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5555555555555556,
|
| 1301 |
+
"grad_norm": 1.895119071006775,
|
| 1302 |
+
"learning_rate": 8.542903020337887e-05,
|
| 1303 |
+
"loss": 0.7014,
|
| 1304 |
+
"step": 185
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5585585585585585,
|
| 1308 |
+
"grad_norm": 2.023733377456665,
|
| 1309 |
+
"learning_rate": 8.449634003358022e-05,
|
| 1310 |
+
"loss": 0.8134,
|
| 1311 |
+
"step": 186
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5615615615615616,
|
| 1315 |
+
"grad_norm": 3.346677303314209,
|
| 1316 |
+
"learning_rate": 8.356502974686941e-05,
|
| 1317 |
+
"loss": 0.8806,
|
| 1318 |
+
"step": 187
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5645645645645646,
|
| 1322 |
+
"grad_norm": 2.9996836185455322,
|
| 1323 |
+
"learning_rate": 8.263518223330697e-05,
|
| 1324 |
+
"loss": 1.3907,
|
| 1325 |
+
"step": 188
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5675675675675675,
|
| 1329 |
+
"grad_norm": 4.624329090118408,
|
| 1330 |
+
"learning_rate": 8.170688025276134e-05,
|
| 1331 |
+
"loss": 1.2675,
|
| 1332 |
+
"step": 189
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.5705705705705706,
|
| 1336 |
+
"grad_norm": 4.569002628326416,
|
| 1337 |
+
"learning_rate": 8.078020642754274e-05,
|
| 1338 |
+
"loss": 1.176,
|
| 1339 |
+
"step": 190
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.5735735735735735,
|
| 1343 |
+
"grad_norm": 4.073513031005859,
|
| 1344 |
+
"learning_rate": 7.985524323504948e-05,
|
| 1345 |
+
"loss": 1.058,
|
| 1346 |
+
"step": 191
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.5765765765765766,
|
| 1350 |
+
"grad_norm": 3.1799652576446533,
|
| 1351 |
+
"learning_rate": 7.89320730004274e-05,
|
| 1352 |
+
"loss": 1.1819,
|
| 1353 |
+
"step": 192
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.5795795795795796,
|
| 1357 |
+
"grad_norm": 3.175107002258301,
|
| 1358 |
+
"learning_rate": 7.801077788924224e-05,
|
| 1359 |
+
"loss": 1.1484,
|
| 1360 |
+
"step": 193
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.5825825825825826,
|
| 1364 |
+
"grad_norm": 7.873251914978027,
|
| 1365 |
+
"learning_rate": 7.709143990016702e-05,
|
| 1366 |
+
"loss": 1.172,
|
| 1367 |
+
"step": 194
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.5855855855855856,
|
| 1371 |
+
"grad_norm": 2.4473395347595215,
|
| 1372 |
+
"learning_rate": 7.617414085768351e-05,
|
| 1373 |
+
"loss": 1.0935,
|
| 1374 |
+
"step": 195
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.5885885885885885,
|
| 1378 |
+
"grad_norm": 2.298689842224121,
|
| 1379 |
+
"learning_rate": 7.525896240479976e-05,
|
| 1380 |
+
"loss": 1.1773,
|
| 1381 |
+
"step": 196
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.5915915915915916,
|
| 1385 |
+
"grad_norm": 3.7667293548583984,
|
| 1386 |
+
"learning_rate": 7.434598599578351e-05,
|
| 1387 |
+
"loss": 1.0643,
|
| 1388 |
+
"step": 197
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.5945945945945946,
|
| 1392 |
+
"grad_norm": 2.9455878734588623,
|
| 1393 |
+
"learning_rate": 7.343529288891239e-05,
|
| 1394 |
+
"loss": 1.227,
|
| 1395 |
+
"step": 198
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.5975975975975976,
|
| 1399 |
+
"grad_norm": 7.777294158935547,
|
| 1400 |
+
"learning_rate": 7.252696413924174e-05,
|
| 1401 |
+
"loss": 1.2713,
|
| 1402 |
+
"step": 199
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.6006006006006006,
|
| 1406 |
+
"grad_norm": 2.827641248703003,
|
| 1407 |
+
"learning_rate": 7.162108059139032e-05,
|
| 1408 |
+
"loss": 1.2447,
|
| 1409 |
+
"step": 200
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6036036036036037,
|
| 1413 |
+
"grad_norm": 2.9091506004333496,
|
| 1414 |
+
"learning_rate": 7.071772287234497e-05,
|
| 1415 |
+
"loss": 1.0227,
|
| 1416 |
+
"step": 201
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.6066066066066066,
|
| 1420 |
+
"grad_norm": 2.9833085536956787,
|
| 1421 |
+
"learning_rate": 6.981697138428434e-05,
|
| 1422 |
+
"loss": 0.9794,
|
| 1423 |
+
"step": 202
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6096096096096096,
|
| 1427 |
+
"grad_norm": 3.3673200607299805,
|
| 1428 |
+
"learning_rate": 6.891890629742288e-05,
|
| 1429 |
+
"loss": 0.9187,
|
| 1430 |
+
"step": 203
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.6126126126126126,
|
| 1434 |
+
"grad_norm": 2.38704252243042,
|
| 1435 |
+
"learning_rate": 6.802360754287547e-05,
|
| 1436 |
+
"loss": 1.0263,
|
| 1437 |
+
"step": 204
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6156156156156156,
|
| 1441 |
+
"grad_norm": 4.426513195037842,
|
| 1442 |
+
"learning_rate": 6.713115480554313e-05,
|
| 1443 |
+
"loss": 1.1677,
|
| 1444 |
+
"step": 205
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.6186186186186187,
|
| 1448 |
+
"grad_norm": 3.6128177642822266,
|
| 1449 |
+
"learning_rate": 6.624162751702076e-05,
|
| 1450 |
+
"loss": 1.3946,
|
| 1451 |
+
"step": 206
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6216216216216216,
|
| 1455 |
+
"grad_norm": 3.2020866870880127,
|
| 1456 |
+
"learning_rate": 6.535510484852767e-05,
|
| 1457 |
+
"loss": 1.5255,
|
| 1458 |
+
"step": 207
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.6246246246246246,
|
| 1462 |
+
"grad_norm": 2.582364082336426,
|
| 1463 |
+
"learning_rate": 6.447166570386063e-05,
|
| 1464 |
+
"loss": 1.1196,
|
| 1465 |
+
"step": 208
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.6276276276276276,
|
| 1469 |
+
"grad_norm": 3.6883230209350586,
|
| 1470 |
+
"learning_rate": 6.35913887123716e-05,
|
| 1471 |
+
"loss": 1.2505,
|
| 1472 |
+
"step": 209
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.6306306306306306,
|
| 1476 |
+
"grad_norm": 2.0863749980926514,
|
| 1477 |
+
"learning_rate": 6.271435222196916e-05,
|
| 1478 |
+
"loss": 0.5439,
|
| 1479 |
+
"step": 210
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.6336336336336337,
|
| 1483 |
+
"grad_norm": 1.9187757968902588,
|
| 1484 |
+
"learning_rate": 6.184063429214515e-05,
|
| 1485 |
+
"loss": 0.6077,
|
| 1486 |
+
"step": 211
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6366366366366366,
|
| 1490 |
+
"grad_norm": 2.5035970211029053,
|
| 1491 |
+
"learning_rate": 6.097031268702746e-05,
|
| 1492 |
+
"loss": 1.0272,
|
| 1493 |
+
"step": 212
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.6396396396396397,
|
| 1497 |
+
"grad_norm": 2.1252877712249756,
|
| 1498 |
+
"learning_rate": 6.010346486845837e-05,
|
| 1499 |
+
"loss": 0.877,
|
| 1500 |
+
"step": 213
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6426426426426426,
|
| 1504 |
+
"grad_norm": 2.2181661128997803,
|
| 1505 |
+
"learning_rate": 5.924016798910037e-05,
|
| 1506 |
+
"loss": 0.7115,
|
| 1507 |
+
"step": 214
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.6456456456456456,
|
| 1511 |
+
"grad_norm": 2.6346538066864014,
|
| 1512 |
+
"learning_rate": 5.838049888556925e-05,
|
| 1513 |
+
"loss": 1.2154,
|
| 1514 |
+
"step": 215
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6486486486486487,
|
| 1518 |
+
"grad_norm": 3.103731870651245,
|
| 1519 |
+
"learning_rate": 5.752453407159522e-05,
|
| 1520 |
+
"loss": 1.021,
|
| 1521 |
+
"step": 216
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.6516516516516516,
|
| 1525 |
+
"grad_norm": 3.415545701980591,
|
| 1526 |
+
"learning_rate": 5.667234973121317e-05,
|
| 1527 |
+
"loss": 1.5522,
|
| 1528 |
+
"step": 217
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.6546546546546547,
|
| 1532 |
+
"grad_norm": 3.22251558303833,
|
| 1533 |
+
"learning_rate": 5.5824021711981686e-05,
|
| 1534 |
+
"loss": 0.9561,
|
| 1535 |
+
"step": 218
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.6576576576576577,
|
| 1539 |
+
"grad_norm": 2.920732259750366,
|
| 1540 |
+
"learning_rate": 5.497962551823266e-05,
|
| 1541 |
+
"loss": 1.0558,
|
| 1542 |
+
"step": 219
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.6606606606606606,
|
| 1546 |
+
"grad_norm": 4.05772590637207,
|
| 1547 |
+
"learning_rate": 5.4139236304350935e-05,
|
| 1548 |
+
"loss": 1.0468,
|
| 1549 |
+
"step": 220
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.6636636636636637,
|
| 1553 |
+
"grad_norm": 2.826936960220337,
|
| 1554 |
+
"learning_rate": 5.33029288680852e-05,
|
| 1555 |
+
"loss": 1.1075,
|
| 1556 |
+
"step": 221
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.6666666666666666,
|
| 1560 |
+
"grad_norm": 4.209110260009766,
|
| 1561 |
+
"learning_rate": 5.247077764389099e-05,
|
| 1562 |
+
"loss": 1.1172,
|
| 1563 |
+
"step": 222
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.6696696696696697,
|
| 1567 |
+
"grad_norm": 3.3651530742645264,
|
| 1568 |
+
"learning_rate": 5.1642856696305575e-05,
|
| 1569 |
+
"loss": 1.0825,
|
| 1570 |
+
"step": 223
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.6726726726726727,
|
| 1574 |
+
"grad_norm": 2.487950086593628,
|
| 1575 |
+
"learning_rate": 5.081923971335582e-05,
|
| 1576 |
+
"loss": 0.7995,
|
| 1577 |
+
"step": 224
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.6756756756756757,
|
| 1581 |
+
"grad_norm": 3.0736727714538574,
|
| 1582 |
+
"learning_rate": 5.000000000000002e-05,
|
| 1583 |
+
"loss": 1.0645,
|
| 1584 |
+
"step": 225
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.6786786786786787,
|
| 1588 |
+
"grad_norm": 4.154182434082031,
|
| 1589 |
+
"learning_rate": 4.918521047160308e-05,
|
| 1590 |
+
"loss": 1.784,
|
| 1591 |
+
"step": 226
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.6816816816816816,
|
| 1595 |
+
"grad_norm": 2.6880645751953125,
|
| 1596 |
+
"learning_rate": 4.837494364744711e-05,
|
| 1597 |
+
"loss": 0.806,
|
| 1598 |
+
"step": 227
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.6846846846846847,
|
| 1602 |
+
"grad_norm": 8.7504301071167,
|
| 1603 |
+
"learning_rate": 4.756927164427685e-05,
|
| 1604 |
+
"loss": 1.2716,
|
| 1605 |
+
"step": 228
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.6876876876876877,
|
| 1609 |
+
"grad_norm": 3.641247510910034,
|
| 1610 |
+
"learning_rate": 4.6768266169880804e-05,
|
| 1611 |
+
"loss": 1.0403,
|
| 1612 |
+
"step": 229
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.6906906906906907,
|
| 1616 |
+
"grad_norm": 2.9611194133758545,
|
| 1617 |
+
"learning_rate": 4.597199851670932e-05,
|
| 1618 |
+
"loss": 1.2985,
|
| 1619 |
+
"step": 230
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.6936936936936937,
|
| 1623 |
+
"grad_norm": 2.1863701343536377,
|
| 1624 |
+
"learning_rate": 4.518053955552903e-05,
|
| 1625 |
+
"loss": 0.9741,
|
| 1626 |
+
"step": 231
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6966966966966966,
|
| 1630 |
+
"grad_norm": 2.5241053104400635,
|
| 1631 |
+
"learning_rate": 4.4393959729115244e-05,
|
| 1632 |
+
"loss": 1.0619,
|
| 1633 |
+
"step": 232
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.6996996996996997,
|
| 1637 |
+
"grad_norm": 2.0366642475128174,
|
| 1638 |
+
"learning_rate": 4.3612329045982236e-05,
|
| 1639 |
+
"loss": 1.0319,
|
| 1640 |
+
"step": 233
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.7027027027027027,
|
| 1644 |
+
"grad_norm": 3.1480021476745605,
|
| 1645 |
+
"learning_rate": 4.283571707415214e-05,
|
| 1646 |
+
"loss": 1.2822,
|
| 1647 |
+
"step": 234
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7057057057057057,
|
| 1651 |
+
"grad_norm": 3.6605899333953857,
|
| 1652 |
+
"learning_rate": 4.206419293496333e-05,
|
| 1653 |
+
"loss": 1.0404,
|
| 1654 |
+
"step": 235
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7087087087087087,
|
| 1658 |
+
"grad_norm": 2.7231175899505615,
|
| 1659 |
+
"learning_rate": 4.129782529691815e-05,
|
| 1660 |
+
"loss": 1.0927,
|
| 1661 |
+
"step": 236
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7117117117117117,
|
| 1665 |
+
"grad_norm": 2.513577938079834,
|
| 1666 |
+
"learning_rate": 4.053668236957134e-05,
|
| 1667 |
+
"loss": 1.2299,
|
| 1668 |
+
"step": 237
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7147147147147147,
|
| 1672 |
+
"grad_norm": 1.9379348754882812,
|
| 1673 |
+
"learning_rate": 3.978083189745907e-05,
|
| 1674 |
+
"loss": 0.7212,
|
| 1675 |
+
"step": 238
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.7177177177177178,
|
| 1679 |
+
"grad_norm": 2.7951018810272217,
|
| 1680 |
+
"learning_rate": 3.903034115406931e-05,
|
| 1681 |
+
"loss": 1.0704,
|
| 1682 |
+
"step": 239
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7207207207207207,
|
| 1686 |
+
"grad_norm": 2.4771158695220947,
|
| 1687 |
+
"learning_rate": 3.828527693585451e-05,
|
| 1688 |
+
"loss": 1.0166,
|
| 1689 |
+
"step": 240
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7237237237237237,
|
| 1693 |
+
"grad_norm": 3.3941121101379395,
|
| 1694 |
+
"learning_rate": 3.7545705556286126e-05,
|
| 1695 |
+
"loss": 1.5398,
|
| 1696 |
+
"step": 241
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7267267267267268,
|
| 1700 |
+
"grad_norm": 2.322334051132202,
|
| 1701 |
+
"learning_rate": 3.681169283995279e-05,
|
| 1702 |
+
"loss": 1.2624,
|
| 1703 |
+
"step": 242
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7297297297297297,
|
| 1707 |
+
"grad_norm": 2.8599162101745605,
|
| 1708 |
+
"learning_rate": 3.6083304116701535e-05,
|
| 1709 |
+
"loss": 1.1381,
|
| 1710 |
+
"step": 243
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7327327327327328,
|
| 1714 |
+
"grad_norm": 4.170224666595459,
|
| 1715 |
+
"learning_rate": 3.536060421582309e-05,
|
| 1716 |
+
"loss": 1.4004,
|
| 1717 |
+
"step": 244
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7357357357357357,
|
| 1721 |
+
"grad_norm": 3.191027879714966,
|
| 1722 |
+
"learning_rate": 3.464365746028208e-05,
|
| 1723 |
+
"loss": 1.1464,
|
| 1724 |
+
"step": 245
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7387387387387387,
|
| 1728 |
+
"grad_norm": 2.834343671798706,
|
| 1729 |
+
"learning_rate": 3.393252766099187e-05,
|
| 1730 |
+
"loss": 1.142,
|
| 1731 |
+
"step": 246
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.7417417417417418,
|
| 1735 |
+
"grad_norm": 3.2168824672698975,
|
| 1736 |
+
"learning_rate": 3.322727811113516e-05,
|
| 1737 |
+
"loss": 1.2666,
|
| 1738 |
+
"step": 247
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7447447447447447,
|
| 1742 |
+
"grad_norm": 2.456047296524048,
|
| 1743 |
+
"learning_rate": 3.252797158053077e-05,
|
| 1744 |
+
"loss": 1.2571,
|
| 1745 |
+
"step": 248
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.7477477477477478,
|
| 1749 |
+
"grad_norm": 2.9790220260620117,
|
| 1750 |
+
"learning_rate": 3.1834670310046734e-05,
|
| 1751 |
+
"loss": 1.2003,
|
| 1752 |
+
"step": 249
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7507507507507507,
|
| 1756 |
+
"grad_norm": 3.3574981689453125,
|
| 1757 |
+
"learning_rate": 3.114743600606078e-05,
|
| 1758 |
+
"loss": 1.3745,
|
| 1759 |
+
"step": 250
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.7537537537537538,
|
| 1763 |
+
"grad_norm": 2.140984296798706,
|
| 1764 |
+
"learning_rate": 3.0466329834968233e-05,
|
| 1765 |
+
"loss": 0.8346,
|
| 1766 |
+
"step": 251
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.7567567567567568,
|
| 1770 |
+
"grad_norm": 3.07039475440979,
|
| 1771 |
+
"learning_rate": 2.979141241773775e-05,
|
| 1772 |
+
"loss": 1.2459,
|
| 1773 |
+
"step": 252
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.7597597597597597,
|
| 1777 |
+
"grad_norm": 2.29309344291687,
|
| 1778 |
+
"learning_rate": 2.9122743824516195e-05,
|
| 1779 |
+
"loss": 1.0559,
|
| 1780 |
+
"step": 253
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.7627627627627628,
|
| 1784 |
+
"grad_norm": 2.319108247756958,
|
| 1785 |
+
"learning_rate": 2.8460383569281824e-05,
|
| 1786 |
+
"loss": 0.8545,
|
| 1787 |
+
"step": 254
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.7657657657657657,
|
| 1791 |
+
"grad_norm": 3.5860605239868164,
|
| 1792 |
+
"learning_rate": 2.7804390604547557e-05,
|
| 1793 |
+
"loss": 0.9624,
|
| 1794 |
+
"step": 255
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.7687687687687688,
|
| 1798 |
+
"grad_norm": 2.4737586975097656,
|
| 1799 |
+
"learning_rate": 2.7154823316113932e-05,
|
| 1800 |
+
"loss": 1.1101,
|
| 1801 |
+
"step": 256
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.7717717717717718,
|
| 1805 |
+
"grad_norm": 2.6330103874206543,
|
| 1806 |
+
"learning_rate": 2.6511739517872426e-05,
|
| 1807 |
+
"loss": 1.1027,
|
| 1808 |
+
"step": 257
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.7747747747747747,
|
| 1812 |
+
"grad_norm": 2.807921886444092,
|
| 1813 |
+
"learning_rate": 2.587519644666001e-05,
|
| 1814 |
+
"loss": 0.8148,
|
| 1815 |
+
"step": 258
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.7777777777777778,
|
| 1819 |
+
"grad_norm": 4.0378546714782715,
|
| 1820 |
+
"learning_rate": 2.5245250757164663e-05,
|
| 1821 |
+
"loss": 1.3036,
|
| 1822 |
+
"step": 259
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.7807807807807807,
|
| 1826 |
+
"grad_norm": 2.8638672828674316,
|
| 1827 |
+
"learning_rate": 2.462195851688306e-05,
|
| 1828 |
+
"loss": 0.8851,
|
| 1829 |
+
"step": 260
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.7837837837837838,
|
| 1833 |
+
"grad_norm": 2.558032512664795,
|
| 1834 |
+
"learning_rate": 2.4005375201130274e-05,
|
| 1835 |
+
"loss": 1.1158,
|
| 1836 |
+
"step": 261
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.7867867867867868,
|
| 1840 |
+
"grad_norm": 1.8975292444229126,
|
| 1841 |
+
"learning_rate": 2.339555568810221e-05,
|
| 1842 |
+
"loss": 0.8933,
|
| 1843 |
+
"step": 262
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.7897897897897898,
|
| 1847 |
+
"grad_norm": 2.8043062686920166,
|
| 1848 |
+
"learning_rate": 2.2792554253991415e-05,
|
| 1849 |
+
"loss": 0.9539,
|
| 1850 |
+
"step": 263
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7927927927927928,
|
| 1854 |
+
"grad_norm": 3.37785267829895,
|
| 1855 |
+
"learning_rate": 2.2196424568156073e-05,
|
| 1856 |
+
"loss": 1.3296,
|
| 1857 |
+
"step": 264
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.7957957957957958,
|
| 1861 |
+
"grad_norm": 2.7959485054016113,
|
| 1862 |
+
"learning_rate": 2.160721968834344e-05,
|
| 1863 |
+
"loss": 1.1462,
|
| 1864 |
+
"step": 265
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.7987987987987988,
|
| 1868 |
+
"grad_norm": 7.449779510498047,
|
| 1869 |
+
"learning_rate": 2.102499205596743e-05,
|
| 1870 |
+
"loss": 1.0893,
|
| 1871 |
+
"step": 266
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8018018018018018,
|
| 1875 |
+
"grad_norm": 6.506124019622803,
|
| 1876 |
+
"learning_rate": 2.0449793491441028e-05,
|
| 1877 |
+
"loss": 1.1367,
|
| 1878 |
+
"step": 267
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8048048048048048,
|
| 1882 |
+
"grad_norm": 2.821166515350342,
|
| 1883 |
+
"learning_rate": 1.9881675189564254e-05,
|
| 1884 |
+
"loss": 1.6805,
|
| 1885 |
+
"step": 268
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8078078078078078,
|
| 1889 |
+
"grad_norm": 2.951826333999634,
|
| 1890 |
+
"learning_rate": 1.93206877149676e-05,
|
| 1891 |
+
"loss": 0.8443,
|
| 1892 |
+
"step": 269
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.8108108108108109,
|
| 1896 |
+
"grad_norm": 3.3361992835998535,
|
| 1897 |
+
"learning_rate": 1.8766880997611424e-05,
|
| 1898 |
+
"loss": 1.2579,
|
| 1899 |
+
"step": 270
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.8138138138138138,
|
| 1903 |
+
"grad_norm": 1.7760874032974243,
|
| 1904 |
+
"learning_rate": 1.8220304328342252e-05,
|
| 1905 |
+
"loss": 0.6704,
|
| 1906 |
+
"step": 271
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.8168168168168168,
|
| 1910 |
+
"grad_norm": 3.165902614593506,
|
| 1911 |
+
"learning_rate": 1.7681006354505493e-05,
|
| 1912 |
+
"loss": 1.0814,
|
| 1913 |
+
"step": 272
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8198198198198198,
|
| 1917 |
+
"grad_norm": 2.863865613937378,
|
| 1918 |
+
"learning_rate": 1.7149035075615794e-05,
|
| 1919 |
+
"loss": 1.0898,
|
| 1920 |
+
"step": 273
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8228228228228228,
|
| 1924 |
+
"grad_norm": 2.5821480751037598,
|
| 1925 |
+
"learning_rate": 1.6624437839084862e-05,
|
| 1926 |
+
"loss": 0.8588,
|
| 1927 |
+
"step": 274
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8258258258258259,
|
| 1931 |
+
"grad_norm": 3.5799527168273926,
|
| 1932 |
+
"learning_rate": 1.6107261336007285e-05,
|
| 1933 |
+
"loss": 1.2082,
|
| 1934 |
+
"step": 275
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.8288288288288288,
|
| 1938 |
+
"grad_norm": 1.9022583961486816,
|
| 1939 |
+
"learning_rate": 1.5597551597004966e-05,
|
| 1940 |
+
"loss": 0.8508,
|
| 1941 |
+
"step": 276
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8318318318318318,
|
| 1945 |
+
"grad_norm": 3.3668415546417236,
|
| 1946 |
+
"learning_rate": 1.5095353988130235e-05,
|
| 1947 |
+
"loss": 1.4462,
|
| 1948 |
+
"step": 277
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8348348348348348,
|
| 1952 |
+
"grad_norm": 3.1614718437194824,
|
| 1953 |
+
"learning_rate": 1.4600713206827932e-05,
|
| 1954 |
+
"loss": 0.7332,
|
| 1955 |
+
"step": 278
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.8378378378378378,
|
| 1959 |
+
"grad_norm": 3.1946263313293457,
|
| 1960 |
+
"learning_rate": 1.4113673277957395e-05,
|
| 1961 |
+
"loss": 1.2424,
|
| 1962 |
+
"step": 279
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.8408408408408409,
|
| 1966 |
+
"grad_norm": 3.346984386444092,
|
| 1967 |
+
"learning_rate": 1.3634277549873953e-05,
|
| 1968 |
+
"loss": 1.1425,
|
| 1969 |
+
"step": 280
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.8438438438438438,
|
| 1973 |
+
"grad_norm": 2.6733968257904053,
|
| 1974 |
+
"learning_rate": 1.3162568690570743e-05,
|
| 1975 |
+
"loss": 1.0175,
|
| 1976 |
+
"step": 281
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.8468468468468469,
|
| 1980 |
+
"grad_norm": 3.606302499771118,
|
| 1981 |
+
"learning_rate": 1.2698588683881186e-05,
|
| 1982 |
+
"loss": 1.1272,
|
| 1983 |
+
"step": 282
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.8498498498498499,
|
| 1987 |
+
"grad_norm": 3.11902117729187,
|
| 1988 |
+
"learning_rate": 1.224237882574224e-05,
|
| 1989 |
+
"loss": 0.8483,
|
| 1990 |
+
"step": 283
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.8528528528528528,
|
| 1994 |
+
"grad_norm": 2.8845720291137695,
|
| 1995 |
+
"learning_rate": 1.1793979720518866e-05,
|
| 1996 |
+
"loss": 0.9861,
|
| 1997 |
+
"step": 284
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.8558558558558559,
|
| 2001 |
+
"grad_norm": 3.02492094039917,
|
| 2002 |
+
"learning_rate": 1.1353431277390126e-05,
|
| 2003 |
+
"loss": 1.2887,
|
| 2004 |
+
"step": 285
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.8588588588588588,
|
| 2008 |
+
"grad_norm": 3.848069906234741,
|
| 2009 |
+
"learning_rate": 1.0920772706797167e-05,
|
| 2010 |
+
"loss": 1.0295,
|
| 2011 |
+
"step": 286
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.8618618618618619,
|
| 2015 |
+
"grad_norm": 2.647904396057129,
|
| 2016 |
+
"learning_rate": 1.0496042516953209e-05,
|
| 2017 |
+
"loss": 0.9551,
|
| 2018 |
+
"step": 287
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.8648648648648649,
|
| 2022 |
+
"grad_norm": 2.807253360748291,
|
| 2023 |
+
"learning_rate": 1.0079278510416313e-05,
|
| 2024 |
+
"loss": 0.7592,
|
| 2025 |
+
"step": 288
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.8678678678678678,
|
| 2029 |
+
"grad_norm": 3.8293511867523193,
|
| 2030 |
+
"learning_rate": 9.670517780724775e-06,
|
| 2031 |
+
"loss": 1.0225,
|
| 2032 |
+
"step": 289
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.8708708708708709,
|
| 2036 |
+
"grad_norm": 2.4294826984405518,
|
| 2037 |
+
"learning_rate": 9.269796709095558e-06,
|
| 2038 |
+
"loss": 1.4176,
|
| 2039 |
+
"step": 290
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.8738738738738738,
|
| 2043 |
+
"grad_norm": 4.781210899353027,
|
| 2044 |
+
"learning_rate": 8.87715096118642e-06,
|
| 2045 |
+
"loss": 0.8798,
|
| 2046 |
+
"step": 291
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.8768768768768769,
|
| 2050 |
+
"grad_norm": 2.406425714492798,
|
| 2051 |
+
"learning_rate": 8.492615483921395e-06,
|
| 2052 |
+
"loss": 0.8694,
|
| 2053 |
+
"step": 292
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.8798798798798799,
|
| 2057 |
+
"grad_norm": 1.9549373388290405,
|
| 2058 |
+
"learning_rate": 8.116224502380387e-06,
|
| 2059 |
+
"loss": 1.1899,
|
| 2060 |
+
"step": 293
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.8828828828828829,
|
| 2064 |
+
"grad_norm": 3.0934197902679443,
|
| 2065 |
+
"learning_rate": 7.74801151675314e-06,
|
| 2066 |
+
"loss": 0.9323,
|
| 2067 |
+
"step": 294
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.8858858858858859,
|
| 2071 |
+
"grad_norm": 4.107941150665283,
|
| 2072 |
+
"learning_rate": 7.3880092993574125e-06,
|
| 2073 |
+
"loss": 1.3923,
|
| 2074 |
+
"step": 295
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.8888888888888888,
|
| 2078 |
+
"grad_norm": 2.7095839977264404,
|
| 2079 |
+
"learning_rate": 7.03624989172228e-06,
|
| 2080 |
+
"loss": 0.8522,
|
| 2081 |
+
"step": 296
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.8918918918918919,
|
| 2085 |
+
"grad_norm": 1.8515009880065918,
|
| 2086 |
+
"learning_rate": 6.692764601736268e-06,
|
| 2087 |
+
"loss": 0.6926,
|
| 2088 |
+
"step": 297
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.8948948948948949,
|
| 2092 |
+
"grad_norm": 3.2831356525421143,
|
| 2093 |
+
"learning_rate": 6.357584000860761e-06,
|
| 2094 |
+
"loss": 0.9239,
|
| 2095 |
+
"step": 298
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.8978978978978979,
|
| 2099 |
+
"grad_norm": 2.5845866203308105,
|
| 2100 |
+
"learning_rate": 6.030737921409169e-06,
|
| 2101 |
+
"loss": 0.8028,
|
| 2102 |
+
"step": 299
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.9009009009009009,
|
| 2106 |
+
"grad_norm": 2.855114221572876,
|
| 2107 |
+
"learning_rate": 5.71225545389158e-06,
|
| 2108 |
+
"loss": 1.1832,
|
| 2109 |
+
"step": 300
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.9039039039039038,
|
| 2113 |
+
"grad_norm": 3.3236331939697266,
|
| 2114 |
+
"learning_rate": 5.402164944425758e-06,
|
| 2115 |
+
"loss": 1.1724,
|
| 2116 |
+
"step": 301
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.9069069069069069,
|
| 2120 |
+
"grad_norm": 6.83448600769043,
|
| 2121 |
+
"learning_rate": 5.100493992214128e-06,
|
| 2122 |
+
"loss": 1.3314,
|
| 2123 |
+
"step": 302
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.9099099099099099,
|
| 2127 |
+
"grad_norm": 2.863325834274292,
|
| 2128 |
+
"learning_rate": 4.807269447087348e-06,
|
| 2129 |
+
"loss": 1.0996,
|
| 2130 |
+
"step": 303
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.9129129129129129,
|
| 2134 |
+
"grad_norm": 3.2937326431274414,
|
| 2135 |
+
"learning_rate": 4.5225174071146455e-06,
|
| 2136 |
+
"loss": 1.1262,
|
| 2137 |
+
"step": 304
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9159159159159159,
|
| 2141 |
+
"grad_norm": 4.532299995422363,
|
| 2142 |
+
"learning_rate": 4.24626321628091e-06,
|
| 2143 |
+
"loss": 0.8499,
|
| 2144 |
+
"step": 305
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.918918918918919,
|
| 2148 |
+
"grad_norm": 3.1992688179016113,
|
| 2149 |
+
"learning_rate": 3.9785314622310495e-06,
|
| 2150 |
+
"loss": 1.3525,
|
| 2151 |
+
"step": 306
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9219219219219219,
|
| 2155 |
+
"grad_norm": 2.4705193042755127,
|
| 2156 |
+
"learning_rate": 3.7193459740815674e-06,
|
| 2157 |
+
"loss": 1.0908,
|
| 2158 |
+
"step": 307
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.924924924924925,
|
| 2162 |
+
"grad_norm": 2.904529333114624,
|
| 2163 |
+
"learning_rate": 3.4687298202996655e-06,
|
| 2164 |
+
"loss": 0.9582,
|
| 2165 |
+
"step": 308
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9279279279279279,
|
| 2169 |
+
"grad_norm": 2.3138301372528076,
|
| 2170 |
+
"learning_rate": 3.226705306650113e-06,
|
| 2171 |
+
"loss": 1.2733,
|
| 2172 |
+
"step": 309
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9309309309309309,
|
| 2176 |
+
"grad_norm": 2.09416127204895,
|
| 2177 |
+
"learning_rate": 2.9932939742099208e-06,
|
| 2178 |
+
"loss": 0.7504,
|
| 2179 |
+
"step": 310
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.933933933933934,
|
| 2183 |
+
"grad_norm": 3.167368173599243,
|
| 2184 |
+
"learning_rate": 2.7685165974510986e-06,
|
| 2185 |
+
"loss": 1.2412,
|
| 2186 |
+
"step": 311
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9369369369369369,
|
| 2190 |
+
"grad_norm": 3.094839572906494,
|
| 2191 |
+
"learning_rate": 2.552393182391677e-06,
|
| 2192 |
+
"loss": 0.9607,
|
| 2193 |
+
"step": 312
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.93993993993994,
|
| 2197 |
+
"grad_norm": 2.5439722537994385,
|
| 2198 |
+
"learning_rate": 2.3449429648150665e-06,
|
| 2199 |
+
"loss": 1.1211,
|
| 2200 |
+
"step": 313
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.9429429429429429,
|
| 2204 |
+
"grad_norm": 3.0354812145233154,
|
| 2205 |
+
"learning_rate": 2.1461844085580385e-06,
|
| 2206 |
+
"loss": 1.1613,
|
| 2207 |
+
"step": 314
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.9459459459459459,
|
| 2211 |
+
"grad_norm": 2.3744139671325684,
|
| 2212 |
+
"learning_rate": 1.9561352038673263e-06,
|
| 2213 |
+
"loss": 0.9695,
|
| 2214 |
+
"step": 315
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 0.948948948948949,
|
| 2218 |
+
"grad_norm": 4.216712474822998,
|
| 2219 |
+
"learning_rate": 1.7748122658251876e-06,
|
| 2220 |
+
"loss": 0.7317,
|
| 2221 |
+
"step": 316
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 0.9519519519519519,
|
| 2225 |
+
"grad_norm": 2.2783091068267822,
|
| 2226 |
+
"learning_rate": 1.6022317328438506e-06,
|
| 2227 |
+
"loss": 1.0139,
|
| 2228 |
+
"step": 317
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 0.954954954954955,
|
| 2232 |
+
"grad_norm": 3.2791147232055664,
|
| 2233 |
+
"learning_rate": 1.4384089652291543e-06,
|
| 2234 |
+
"loss": 1.1532,
|
| 2235 |
+
"step": 318
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 0.9579579579579579,
|
| 2239 |
+
"grad_norm": 2.5588083267211914,
|
| 2240 |
+
"learning_rate": 1.2833585438134287e-06,
|
| 2241 |
+
"loss": 0.9971,
|
| 2242 |
+
"step": 319
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.960960960960961,
|
| 2246 |
+
"grad_norm": 2.4663734436035156,
|
| 2247 |
+
"learning_rate": 1.1370942686577347e-06,
|
| 2248 |
+
"loss": 1.1583,
|
| 2249 |
+
"step": 320
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 0.963963963963964,
|
| 2253 |
+
"grad_norm": 3.0910394191741943,
|
| 2254 |
+
"learning_rate": 9.996291578236228e-07,
|
| 2255 |
+
"loss": 1.3676,
|
| 2256 |
+
"step": 321
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 0.9669669669669669,
|
| 2260 |
+
"grad_norm": 2.7286369800567627,
|
| 2261 |
+
"learning_rate": 8.709754462144615e-07,
|
| 2262 |
+
"loss": 1.1954,
|
| 2263 |
+
"step": 322
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 0.96996996996997,
|
| 2267 |
+
"grad_norm": 2.4788362979888916,
|
| 2268 |
+
"learning_rate": 7.511445844864962e-07,
|
| 2269 |
+
"loss": 0.6394,
|
| 2270 |
+
"step": 323
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 0.972972972972973,
|
| 2274 |
+
"grad_norm": 2.686504364013672,
|
| 2275 |
+
"learning_rate": 6.401472380297091e-07,
|
| 2276 |
+
"loss": 1.2232,
|
| 2277 |
+
"step": 324
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 0.975975975975976,
|
| 2281 |
+
"grad_norm": 2.3043410778045654,
|
| 2282 |
+
"learning_rate": 5.379932860185122e-07,
|
| 2283 |
+
"loss": 0.8467,
|
| 2284 |
+
"step": 325
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 0.978978978978979,
|
| 2288 |
+
"grad_norm": 2.39601993560791,
|
| 2289 |
+
"learning_rate": 4.44691820532539e-07,
|
| 2290 |
+
"loss": 1.031,
|
| 2291 |
+
"step": 326
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 0.9819819819819819,
|
| 2295 |
+
"grad_norm": 3.4683892726898193,
|
| 2296 |
+
"learning_rate": 3.6025114574734785e-07,
|
| 2297 |
+
"loss": 1.2397,
|
| 2298 |
+
"step": 327
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.984984984984985,
|
| 2302 |
+
"grad_norm": 3.130098819732666,
|
| 2303 |
+
"learning_rate": 2.846787771953574e-07,
|
| 2304 |
+
"loss": 1.0141,
|
| 2305 |
+
"step": 328
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 0.987987987987988,
|
| 2309 |
+
"grad_norm": 3.894436836242676,
|
| 2310 |
+
"learning_rate": 2.179814410969261e-07,
|
| 2311 |
+
"loss": 1.4286,
|
| 2312 |
+
"step": 329
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 0.990990990990991,
|
| 2316 |
+
"grad_norm": 3.297236919403076,
|
| 2317 |
+
"learning_rate": 1.6016507376169777e-07,
|
| 2318 |
+
"loss": 1.2933,
|
| 2319 |
+
"step": 330
|
| 2320 |
+
},
|
| 2321 |
+
{
|
| 2322 |
+
"epoch": 0.993993993993994,
|
| 2323 |
+
"grad_norm": 4.986927509307861,
|
| 2324 |
+
"learning_rate": 1.1123482106021322e-07,
|
| 2325 |
+
"loss": 1.6548,
|
| 2326 |
+
"step": 331
|
| 2327 |
+
},
|
| 2328 |
+
{
|
| 2329 |
+
"epoch": 0.996996996996997,
|
| 2330 |
+
"grad_norm": 3.8665215969085693,
|
| 2331 |
+
"learning_rate": 7.119503796599868e-08,
|
| 2332 |
+
"loss": 0.8686,
|
| 2333 |
+
"step": 332
|
| 2334 |
+
},
|
| 2335 |
+
{
|
| 2336 |
+
"epoch": 1.0,
|
| 2337 |
+
"grad_norm": 2.562675952911377,
|
| 2338 |
+
"learning_rate": 4.0049288167842705e-08,
|
| 2339 |
+
"loss": 0.8482,
|
| 2340 |
+
"step": 333
|
| 2341 |
+
},
|
| 2342 |
+
{
|
| 2343 |
+
"epoch": 1.0,
|
| 2344 |
+
"eval_loss": 1.0635182857513428,
|
| 2345 |
+
"eval_runtime": 6.3055,
|
| 2346 |
+
"eval_samples_per_second": 11.26,
|
| 2347 |
+
"eval_steps_per_second": 11.26,
|
| 2348 |
+
"step": 333
|
| 2349 |
}
|
| 2350 |
],
|
| 2351 |
"logging_steps": 1,
|
| 2352 |
+
"max_steps": 333,
|
| 2353 |
"num_input_tokens_seen": 0,
|
| 2354 |
"num_train_epochs": 1,
|
| 2355 |
+
"save_steps": 500,
|
| 2356 |
"stateful_callbacks": {
|
| 2357 |
"TrainerControl": {
|
| 2358 |
"args": {
|
|
|
|
| 2365 |
"attributes": {}
|
| 2366 |
}
|
| 2367 |
},
|
| 2368 |
+
"total_flos": 6.159048946325914e+16,
|
| 2369 |
"train_batch_size": 1,
|
| 2370 |
"trial_name": null,
|
| 2371 |
"trial_params": null
|
last-checkpoint/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6776
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd5b5e9829e2746865387002c1cb7d44db56f890fcf35267b4f105596244cfe5
|
| 3 |
size 6776
|