MHGanainy/gpt2-xl-lora-multi-512-k5-7-im-0
Browse files- README.md +2 -0
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Oct22_13-46-58_6a6876ac5cbf/events.out.tfevents.1729608505.6a6876ac5cbf.6998.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +721 -0
README.md
CHANGED
|
@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
# gpt2-xl-lora-multi-512-k5-7-im-0
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [openai-community/gpt2-xl](https://huggingface.co/openai-community/gpt2-xl) on an unknown dataset.
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
|
|
| 15 |
# gpt2-xl-lora-multi-512-k5-7-im-0
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [openai-community/gpt2-xl](https://huggingface.co/openai-community/gpt2-xl) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 1.8678
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 1.867760181427002,
|
| 4 |
+
"eval_runtime": 104.988,
|
| 5 |
+
"eval_samples_per_second": 145.788,
|
| 6 |
+
"eval_steps_per_second": 9.115,
|
| 7 |
+
"perplexity": 6.4737800555614,
|
| 8 |
+
"total_flos": 1.421397042852266e+18,
|
| 9 |
+
"train_loss": 2.0159098046724915,
|
| 10 |
+
"train_runtime": 3573.2428,
|
| 11 |
+
"train_samples_per_second": 43.7,
|
| 12 |
+
"train_steps_per_second": 2.731
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 1.867760181427002,
|
| 4 |
+
"eval_runtime": 104.988,
|
| 5 |
+
"eval_samples_per_second": 145.788,
|
| 6 |
+
"eval_steps_per_second": 9.115,
|
| 7 |
+
"perplexity": 6.4737800555614
|
| 8 |
+
}
|
runs/Oct22_13-46-58_6a6876ac5cbf/events.out.tfevents.1729608505.6a6876ac5cbf.6998.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2c6f34aaa545b4af135e990d2bbd06d11c37234f3e0ab82b13547327c8368ab1
|
| 3 |
+
size 311
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.421397042852266e+18,
|
| 4 |
+
"train_loss": 2.0159098046724915,
|
| 5 |
+
"train_runtime": 3573.2428,
|
| 6 |
+
"train_samples_per_second": 43.7,
|
| 7 |
+
"train_steps_per_second": 2.731
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,721 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 9760,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.010245901639344262,
|
| 13 |
+
"grad_norm": 0.055161599069833755,
|
| 14 |
+
"learning_rate": 2.5618035096708087e-07,
|
| 15 |
+
"loss": 2.2315,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.020491803278688523,
|
| 20 |
+
"grad_norm": 0.05540880188345909,
|
| 21 |
+
"learning_rate": 5.123607019341617e-07,
|
| 22 |
+
"loss": 2.2296,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.030737704918032786,
|
| 27 |
+
"grad_norm": 0.0638064593076706,
|
| 28 |
+
"learning_rate": 7.685410529012426e-07,
|
| 29 |
+
"loss": 2.237,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.040983606557377046,
|
| 34 |
+
"grad_norm": 0.0683213621377945,
|
| 35 |
+
"learning_rate": 1.0247214038683235e-06,
|
| 36 |
+
"loss": 2.2204,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.05122950819672131,
|
| 41 |
+
"grad_norm": 0.08443646132946014,
|
| 42 |
+
"learning_rate": 1.2809017548354041e-06,
|
| 43 |
+
"loss": 2.2304,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.06147540983606557,
|
| 48 |
+
"grad_norm": 0.07605498284101486,
|
| 49 |
+
"learning_rate": 1.5370821058024852e-06,
|
| 50 |
+
"loss": 2.2239,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.07172131147540983,
|
| 55 |
+
"grad_norm": 0.08266434073448181,
|
| 56 |
+
"learning_rate": 1.7932624567695659e-06,
|
| 57 |
+
"loss": 2.2219,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.08196721311475409,
|
| 62 |
+
"grad_norm": 0.10109508037567139,
|
| 63 |
+
"learning_rate": 2.049442807736647e-06,
|
| 64 |
+
"loss": 2.2004,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.09221311475409837,
|
| 69 |
+
"grad_norm": 0.11055849492549896,
|
| 70 |
+
"learning_rate": 2.3056231587037276e-06,
|
| 71 |
+
"loss": 2.1945,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.10245901639344263,
|
| 76 |
+
"grad_norm": 0.1108393445611,
|
| 77 |
+
"learning_rate": 2.5618035096708083e-06,
|
| 78 |
+
"loss": 2.1769,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.11270491803278689,
|
| 83 |
+
"grad_norm": 0.14474108815193176,
|
| 84 |
+
"learning_rate": 2.8179838606378894e-06,
|
| 85 |
+
"loss": 2.1904,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.12295081967213115,
|
| 90 |
+
"grad_norm": 0.13787466287612915,
|
| 91 |
+
"learning_rate": 3.0741642116049704e-06,
|
| 92 |
+
"loss": 2.1765,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.13319672131147542,
|
| 97 |
+
"grad_norm": 0.14868797361850739,
|
| 98 |
+
"learning_rate": 3.330344562572051e-06,
|
| 99 |
+
"loss": 2.1609,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.14344262295081966,
|
| 104 |
+
"grad_norm": 0.15633544325828552,
|
| 105 |
+
"learning_rate": 3.5865249135391318e-06,
|
| 106 |
+
"loss": 2.1708,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.15368852459016394,
|
| 111 |
+
"grad_norm": 0.17003007233142853,
|
| 112 |
+
"learning_rate": 3.842705264506213e-06,
|
| 113 |
+
"loss": 2.1521,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.16393442622950818,
|
| 118 |
+
"grad_norm": 0.1722685843706131,
|
| 119 |
+
"learning_rate": 4.098885615473294e-06,
|
| 120 |
+
"loss": 2.1417,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.17418032786885246,
|
| 125 |
+
"grad_norm": 0.18862290680408478,
|
| 126 |
+
"learning_rate": 4.3550659664403746e-06,
|
| 127 |
+
"loss": 2.1328,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.18442622950819673,
|
| 132 |
+
"grad_norm": 0.211067795753479,
|
| 133 |
+
"learning_rate": 4.611246317407455e-06,
|
| 134 |
+
"loss": 2.1222,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.19467213114754098,
|
| 139 |
+
"grad_norm": 0.21424412727355957,
|
| 140 |
+
"learning_rate": 4.867426668374536e-06,
|
| 141 |
+
"loss": 2.1448,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.20491803278688525,
|
| 146 |
+
"grad_norm": 0.21230654418468475,
|
| 147 |
+
"learning_rate": 5.1236070193416165e-06,
|
| 148 |
+
"loss": 2.1189,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2151639344262295,
|
| 153 |
+
"grad_norm": 0.2188699096441269,
|
| 154 |
+
"learning_rate": 5.379787370308698e-06,
|
| 155 |
+
"loss": 2.1208,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.22540983606557377,
|
| 160 |
+
"grad_norm": 0.23276151716709137,
|
| 161 |
+
"learning_rate": 5.635967721275779e-06,
|
| 162 |
+
"loss": 2.0994,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.23565573770491804,
|
| 167 |
+
"grad_norm": 0.25028660893440247,
|
| 168 |
+
"learning_rate": 5.892148072242859e-06,
|
| 169 |
+
"loss": 2.0921,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.2459016393442623,
|
| 174 |
+
"grad_norm": 0.23129107058048248,
|
| 175 |
+
"learning_rate": 6.148328423209941e-06,
|
| 176 |
+
"loss": 2.0986,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.25614754098360654,
|
| 181 |
+
"grad_norm": 0.30324786901474,
|
| 182 |
+
"learning_rate": 6.404508774177021e-06,
|
| 183 |
+
"loss": 2.0998,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.26639344262295084,
|
| 188 |
+
"grad_norm": 0.3137739896774292,
|
| 189 |
+
"learning_rate": 6.660689125144102e-06,
|
| 190 |
+
"loss": 2.0831,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.2766393442622951,
|
| 195 |
+
"grad_norm": 0.29233041405677795,
|
| 196 |
+
"learning_rate": 6.916869476111183e-06,
|
| 197 |
+
"loss": 2.072,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.28688524590163933,
|
| 202 |
+
"grad_norm": 0.2947608232498169,
|
| 203 |
+
"learning_rate": 7.1730498270782635e-06,
|
| 204 |
+
"loss": 2.0713,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.29713114754098363,
|
| 209 |
+
"grad_norm": 0.2987135350704193,
|
| 210 |
+
"learning_rate": 7.429230178045345e-06,
|
| 211 |
+
"loss": 2.0745,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.3073770491803279,
|
| 216 |
+
"grad_norm": 0.3173224627971649,
|
| 217 |
+
"learning_rate": 7.685410529012427e-06,
|
| 218 |
+
"loss": 2.0552,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.3176229508196721,
|
| 223 |
+
"grad_norm": 0.3080308139324188,
|
| 224 |
+
"learning_rate": 7.941590879979506e-06,
|
| 225 |
+
"loss": 2.058,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.32786885245901637,
|
| 230 |
+
"grad_norm": 0.3344276249408722,
|
| 231 |
+
"learning_rate": 8.197771230946588e-06,
|
| 232 |
+
"loss": 2.0335,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.33811475409836067,
|
| 237 |
+
"grad_norm": 0.35821136832237244,
|
| 238 |
+
"learning_rate": 8.453951581913668e-06,
|
| 239 |
+
"loss": 2.0389,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3483606557377049,
|
| 244 |
+
"grad_norm": 0.34500202536582947,
|
| 245 |
+
"learning_rate": 8.710131932880749e-06,
|
| 246 |
+
"loss": 2.0473,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.35860655737704916,
|
| 251 |
+
"grad_norm": 0.3924405872821808,
|
| 252 |
+
"learning_rate": 8.96631228384783e-06,
|
| 253 |
+
"loss": 2.0413,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.36885245901639346,
|
| 258 |
+
"grad_norm": 0.3596484065055847,
|
| 259 |
+
"learning_rate": 9.22249263481491e-06,
|
| 260 |
+
"loss": 2.0408,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.3790983606557377,
|
| 265 |
+
"grad_norm": 0.3991853892803192,
|
| 266 |
+
"learning_rate": 9.478672985781992e-06,
|
| 267 |
+
"loss": 2.0143,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.38934426229508196,
|
| 272 |
+
"grad_norm": 0.36749976873397827,
|
| 273 |
+
"learning_rate": 9.734853336749072e-06,
|
| 274 |
+
"loss": 2.0248,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.39959016393442626,
|
| 279 |
+
"grad_norm": 0.41294941306114197,
|
| 280 |
+
"learning_rate": 9.991033687716153e-06,
|
| 281 |
+
"loss": 2.0214,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.4098360655737705,
|
| 286 |
+
"grad_norm": 0.40261194109916687,
|
| 287 |
+
"learning_rate": 1.0247214038683233e-05,
|
| 288 |
+
"loss": 2.0089,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.42008196721311475,
|
| 293 |
+
"grad_norm": 0.3836148977279663,
|
| 294 |
+
"learning_rate": 1.0503394389650315e-05,
|
| 295 |
+
"loss": 2.0109,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.430327868852459,
|
| 300 |
+
"grad_norm": 0.38613057136535645,
|
| 301 |
+
"learning_rate": 1.0759574740617396e-05,
|
| 302 |
+
"loss": 1.9881,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.4405737704918033,
|
| 307 |
+
"grad_norm": 0.39355090260505676,
|
| 308 |
+
"learning_rate": 1.1015755091584478e-05,
|
| 309 |
+
"loss": 1.9919,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.45081967213114754,
|
| 314 |
+
"grad_norm": 0.5257683992385864,
|
| 315 |
+
"learning_rate": 1.1271935442551557e-05,
|
| 316 |
+
"loss": 1.9929,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.4610655737704918,
|
| 321 |
+
"grad_norm": 0.43797537684440613,
|
| 322 |
+
"learning_rate": 1.1528115793518637e-05,
|
| 323 |
+
"loss": 2.0092,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.4713114754098361,
|
| 328 |
+
"grad_norm": 0.5497995018959045,
|
| 329 |
+
"learning_rate": 1.1784296144485719e-05,
|
| 330 |
+
"loss": 1.9967,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.48155737704918034,
|
| 335 |
+
"grad_norm": 0.4768752157688141,
|
| 336 |
+
"learning_rate": 1.20404764954528e-05,
|
| 337 |
+
"loss": 1.9921,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.4918032786885246,
|
| 342 |
+
"grad_norm": 0.43218129873275757,
|
| 343 |
+
"learning_rate": 1.2296656846419882e-05,
|
| 344 |
+
"loss": 1.9883,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.5020491803278688,
|
| 349 |
+
"grad_norm": 0.47497090697288513,
|
| 350 |
+
"learning_rate": 1.2552837197386962e-05,
|
| 351 |
+
"loss": 1.9799,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.5122950819672131,
|
| 356 |
+
"grad_norm": 0.4906615912914276,
|
| 357 |
+
"learning_rate": 1.2809017548354041e-05,
|
| 358 |
+
"loss": 1.9762,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.5225409836065574,
|
| 363 |
+
"grad_norm": 0.5236986875534058,
|
| 364 |
+
"learning_rate": 1.3065197899321123e-05,
|
| 365 |
+
"loss": 1.9736,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.5327868852459017,
|
| 370 |
+
"grad_norm": 0.5295742750167847,
|
| 371 |
+
"learning_rate": 1.3318816446778533e-05,
|
| 372 |
+
"loss": 1.9762,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.5430327868852459,
|
| 377 |
+
"grad_norm": 0.4924796521663666,
|
| 378 |
+
"learning_rate": 1.3574996797745615e-05,
|
| 379 |
+
"loss": 1.9709,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.5532786885245902,
|
| 384 |
+
"grad_norm": 0.5359669327735901,
|
| 385 |
+
"learning_rate": 1.3831177148712696e-05,
|
| 386 |
+
"loss": 1.9594,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.5635245901639344,
|
| 391 |
+
"grad_norm": 0.49581173062324524,
|
| 392 |
+
"learning_rate": 1.4087357499679775e-05,
|
| 393 |
+
"loss": 1.9787,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.5737704918032787,
|
| 398 |
+
"grad_norm": 0.49295076727867126,
|
| 399 |
+
"learning_rate": 1.4343537850646856e-05,
|
| 400 |
+
"loss": 1.9628,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.5840163934426229,
|
| 405 |
+
"grad_norm": 0.6119683980941772,
|
| 406 |
+
"learning_rate": 1.4599718201613938e-05,
|
| 407 |
+
"loss": 1.9622,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.5942622950819673,
|
| 412 |
+
"grad_norm": 0.5446974039077759,
|
| 413 |
+
"learning_rate": 1.4855898552581019e-05,
|
| 414 |
+
"loss": 1.9547,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.6045081967213115,
|
| 419 |
+
"grad_norm": 0.5105884671211243,
|
| 420 |
+
"learning_rate": 1.5112078903548099e-05,
|
| 421 |
+
"loss": 1.9477,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.6147540983606558,
|
| 426 |
+
"grad_norm": 0.5642023682594299,
|
| 427 |
+
"learning_rate": 1.536825925451518e-05,
|
| 428 |
+
"loss": 1.9433,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.625,
|
| 433 |
+
"grad_norm": 0.4802306294441223,
|
| 434 |
+
"learning_rate": 1.562443960548226e-05,
|
| 435 |
+
"loss": 1.984,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.6352459016393442,
|
| 440 |
+
"grad_norm": 0.5581754446029663,
|
| 441 |
+
"learning_rate": 1.588061995644934e-05,
|
| 442 |
+
"loss": 1.9341,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.6454918032786885,
|
| 447 |
+
"grad_norm": 0.5883158445358276,
|
| 448 |
+
"learning_rate": 1.6136800307416423e-05,
|
| 449 |
+
"loss": 1.9394,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.6557377049180327,
|
| 454 |
+
"grad_norm": 0.5291092395782471,
|
| 455 |
+
"learning_rate": 1.6392980658383503e-05,
|
| 456 |
+
"loss": 1.9421,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.6659836065573771,
|
| 461 |
+
"grad_norm": 0.6235113739967346,
|
| 462 |
+
"learning_rate": 1.6649161009350586e-05,
|
| 463 |
+
"loss": 1.9602,
|
| 464 |
+
"step": 6500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.6762295081967213,
|
| 468 |
+
"grad_norm": 0.49288830161094666,
|
| 469 |
+
"learning_rate": 1.6905341360317663e-05,
|
| 470 |
+
"loss": 1.9346,
|
| 471 |
+
"step": 6600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.6864754098360656,
|
| 475 |
+
"grad_norm": 0.6099859476089478,
|
| 476 |
+
"learning_rate": 1.7161521711284746e-05,
|
| 477 |
+
"loss": 1.9431,
|
| 478 |
+
"step": 6700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.6967213114754098,
|
| 482 |
+
"grad_norm": 0.5682905316352844,
|
| 483 |
+
"learning_rate": 1.7417702062251826e-05,
|
| 484 |
+
"loss": 1.9137,
|
| 485 |
+
"step": 6800
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.7069672131147541,
|
| 489 |
+
"grad_norm": 0.7166417241096497,
|
| 490 |
+
"learning_rate": 1.767388241321891e-05,
|
| 491 |
+
"loss": 1.9391,
|
| 492 |
+
"step": 6900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.7172131147540983,
|
| 496 |
+
"grad_norm": 0.5477021336555481,
|
| 497 |
+
"learning_rate": 1.793006276418599e-05,
|
| 498 |
+
"loss": 1.9429,
|
| 499 |
+
"step": 7000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.7274590163934426,
|
| 503 |
+
"grad_norm": 0.54074627161026,
|
| 504 |
+
"learning_rate": 1.818624311515307e-05,
|
| 505 |
+
"loss": 1.9384,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.7377049180327869,
|
| 510 |
+
"grad_norm": 0.5290340185165405,
|
| 511 |
+
"learning_rate": 1.8442423466120148e-05,
|
| 512 |
+
"loss": 1.9356,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.7479508196721312,
|
| 517 |
+
"grad_norm": 0.5777207612991333,
|
| 518 |
+
"learning_rate": 1.869860381708723e-05,
|
| 519 |
+
"loss": 1.9184,
|
| 520 |
+
"step": 7300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.7581967213114754,
|
| 524 |
+
"grad_norm": 0.6420373320579529,
|
| 525 |
+
"learning_rate": 1.895478416805431e-05,
|
| 526 |
+
"loss": 1.9236,
|
| 527 |
+
"step": 7400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.7684426229508197,
|
| 531 |
+
"grad_norm": 0.5742615461349487,
|
| 532 |
+
"learning_rate": 1.9210964519021394e-05,
|
| 533 |
+
"loss": 1.9222,
|
| 534 |
+
"step": 7500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.7786885245901639,
|
| 538 |
+
"grad_norm": 0.613657534122467,
|
| 539 |
+
"learning_rate": 1.946714486998847e-05,
|
| 540 |
+
"loss": 1.9362,
|
| 541 |
+
"step": 7600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.7889344262295082,
|
| 545 |
+
"grad_norm": 0.5610764622688293,
|
| 546 |
+
"learning_rate": 1.9723325220955554e-05,
|
| 547 |
+
"loss": 1.9035,
|
| 548 |
+
"step": 7700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.7991803278688525,
|
| 552 |
+
"grad_norm": 0.6575483083724976,
|
| 553 |
+
"learning_rate": 1.9979505571922634e-05,
|
| 554 |
+
"loss": 1.9312,
|
| 555 |
+
"step": 7800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.8094262295081968,
|
| 559 |
+
"grad_norm": 0.5548086166381836,
|
| 560 |
+
"learning_rate": 1.9890692959274893e-05,
|
| 561 |
+
"loss": 1.9221,
|
| 562 |
+
"step": 7900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.819672131147541,
|
| 566 |
+
"grad_norm": 0.6601373553276062,
|
| 567 |
+
"learning_rate": 1.952683486776797e-05,
|
| 568 |
+
"loss": 1.9076,
|
| 569 |
+
"step": 8000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.8299180327868853,
|
| 573 |
+
"grad_norm": 0.5061026215553284,
|
| 574 |
+
"learning_rate": 1.891699257502426e-05,
|
| 575 |
+
"loss": 1.9173,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.8401639344262295,
|
| 580 |
+
"grad_norm": 0.5650314688682556,
|
| 581 |
+
"learning_rate": 1.807691229385665e-05,
|
| 582 |
+
"loss": 1.9175,
|
| 583 |
+
"step": 8200
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.8504098360655737,
|
| 587 |
+
"grad_norm": 0.7669892311096191,
|
| 588 |
+
"learning_rate": 1.7028285013959614e-05,
|
| 589 |
+
"loss": 1.9074,
|
| 590 |
+
"step": 8300
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.860655737704918,
|
| 594 |
+
"grad_norm": 0.6210095286369324,
|
| 595 |
+
"learning_rate": 1.5798186437538642e-05,
|
| 596 |
+
"loss": 1.9109,
|
| 597 |
+
"step": 8400
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.8709016393442623,
|
| 601 |
+
"grad_norm": 0.6777961254119873,
|
| 602 |
+
"learning_rate": 1.4418377880913716e-05,
|
| 603 |
+
"loss": 1.9079,
|
| 604 |
+
"step": 8500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.8811475409836066,
|
| 608 |
+
"grad_norm": 0.5773517489433289,
|
| 609 |
+
"learning_rate": 1.2924486192914705e-05,
|
| 610 |
+
"loss": 1.9298,
|
| 611 |
+
"step": 8600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.8913934426229508,
|
| 615 |
+
"grad_norm": 0.5826138257980347,
|
| 616 |
+
"learning_rate": 1.1355083864689736e-05,
|
| 617 |
+
"loss": 1.9094,
|
| 618 |
+
"step": 8700
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.9016393442622951,
|
| 622 |
+
"grad_norm": 0.6318073868751526,
|
| 623 |
+
"learning_rate": 9.750693082619274e-06,
|
| 624 |
+
"loss": 1.9103,
|
| 625 |
+
"step": 8800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.9118852459016393,
|
| 629 |
+
"grad_norm": 0.6698086261749268,
|
| 630 |
+
"learning_rate": 8.15273943982811e-06,
|
| 631 |
+
"loss": 1.9159,
|
| 632 |
+
"step": 8900
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.9221311475409836,
|
| 636 |
+
"grad_norm": 0.612259566783905,
|
| 637 |
+
"learning_rate": 6.602482321609596e-06,
|
| 638 |
+
"loss": 1.9195,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.9323770491803278,
|
| 643 |
+
"grad_norm": 0.5766026973724365,
|
| 644 |
+
"learning_rate": 5.139949582359195e-06,
|
| 645 |
+
"loss": 1.9105,
|
| 646 |
+
"step": 9100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.9426229508196722,
|
| 650 |
+
"grad_norm": 0.6413909196853638,
|
| 651 |
+
"learning_rate": 3.8029040208070833e-06,
|
| 652 |
+
"loss": 1.8875,
|
| 653 |
+
"step": 9200
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.9528688524590164,
|
| 657 |
+
"grad_norm": 0.651620090007782,
|
| 658 |
+
"learning_rate": 2.6258683393030516e-06,
|
| 659 |
+
"loss": 1.9158,
|
| 660 |
+
"step": 9300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.9631147540983607,
|
| 664 |
+
"grad_norm": 0.6111190319061279,
|
| 665 |
+
"learning_rate": 1.6392337628405675e-06,
|
| 666 |
+
"loss": 1.9242,
|
| 667 |
+
"step": 9400
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.9733606557377049,
|
| 671 |
+
"grad_norm": 0.6417702436447144,
|
| 672 |
+
"learning_rate": 8.68475333400769e-07,
|
| 673 |
+
"loss": 1.891,
|
| 674 |
+
"step": 9500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.9836065573770492,
|
| 678 |
+
"grad_norm": 0.5989698767662048,
|
| 679 |
+
"learning_rate": 3.3762624823906577e-07,
|
| 680 |
+
"loss": 1.9093,
|
| 681 |
+
"step": 9600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.9938524590163934,
|
| 685 |
+
"grad_norm": 0.7152329683303833,
|
| 686 |
+
"learning_rate": 4.9692246345985905e-08,
|
| 687 |
+
"loss": 1.8982,
|
| 688 |
+
"step": 9700
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.0,
|
| 692 |
+
"step": 9760,
|
| 693 |
+
"total_flos": 1.421397042852266e+18,
|
| 694 |
+
"train_loss": 2.0159098046724915,
|
| 695 |
+
"train_runtime": 3573.2428,
|
| 696 |
+
"train_samples_per_second": 43.7,
|
| 697 |
+
"train_steps_per_second": 2.731
|
| 698 |
+
}
|
| 699 |
+
],
|
| 700 |
+
"logging_steps": 100,
|
| 701 |
+
"max_steps": 9760,
|
| 702 |
+
"num_input_tokens_seen": 0,
|
| 703 |
+
"num_train_epochs": 1,
|
| 704 |
+
"save_steps": 500,
|
| 705 |
+
"stateful_callbacks": {
|
| 706 |
+
"TrainerControl": {
|
| 707 |
+
"args": {
|
| 708 |
+
"should_epoch_stop": false,
|
| 709 |
+
"should_evaluate": false,
|
| 710 |
+
"should_log": false,
|
| 711 |
+
"should_save": true,
|
| 712 |
+
"should_training_stop": true
|
| 713 |
+
},
|
| 714 |
+
"attributes": {}
|
| 715 |
+
}
|
| 716 |
+
},
|
| 717 |
+
"total_flos": 1.421397042852266e+18,
|
| 718 |
+
"train_batch_size": 2,
|
| 719 |
+
"trial_name": null,
|
| 720 |
+
"trial_params": null
|
| 721 |
+
}
|