Training in progress, step 2000, checkpoint
Browse files- checkpoint-2000/config.json +27 -0
- checkpoint-2000/generation_config.json +6 -0
- checkpoint-2000/global_step2000/mp_rank_00_model_states.pt +3 -0
- checkpoint-2000/latest +1 -0
- checkpoint-2000/pytorch_model.bin +3 -0
- checkpoint-2000/rng_state_0.pth +3 -0
- checkpoint-2000/rng_state_1.pth +3 -0
- checkpoint-2000/rng_state_2.pth +3 -0
- checkpoint-2000/rng_state_3.pth +3 -0
- checkpoint-2000/rng_state_4.pth +3 -0
- checkpoint-2000/rng_state_5.pth +3 -0
- checkpoint-2000/rng_state_6.pth +3 -0
- checkpoint-2000/rng_state_7.pth +3 -0
- checkpoint-2000/scheduler.pt +3 -0
- checkpoint-2000/special_tokens_map.json +24 -0
- checkpoint-2000/tokenizer.json +0 -0
- checkpoint-2000/tokenizer_config.json +78 -0
- checkpoint-2000/trainer_state.json +1219 -0
- checkpoint-2000/training_args.bin +3 -0
checkpoint-2000/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "nickypro/tinyllama-110M",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 2048,
|
| 13 |
+
"max_position_embeddings": 1024,
|
| 14 |
+
"model_type": "llama",
|
| 15 |
+
"num_attention_heads": 12,
|
| 16 |
+
"num_hidden_layers": 12,
|
| 17 |
+
"num_key_value_heads": 12,
|
| 18 |
+
"pretraining_tp": 1,
|
| 19 |
+
"rms_norm_eps": 1e-05,
|
| 20 |
+
"rope_scaling": null,
|
| 21 |
+
"rope_theta": 10000.0,
|
| 22 |
+
"tie_word_embeddings": true,
|
| 23 |
+
"torch_dtype": "float16",
|
| 24 |
+
"transformers_version": "4.34.1",
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 46336
|
| 27 |
+
}
|
checkpoint-2000/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.34.1"
|
| 6 |
+
}
|
checkpoint-2000/global_step2000/mp_rank_00_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7212a9551d0c75d80bdc61705849de92d222c3750909cc28ce955d352d5b5f71
|
| 3 |
+
size 1687740870
|
checkpoint-2000/latest
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
global_step2000
|
checkpoint-2000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f8c1faed6d04c84b3bdbda0e20d51a3631f4d4b453183656d2877a322edb40e8
|
| 3 |
+
size 312287814
|
checkpoint-2000/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0f8f81f81a21cdf7d6a9d642f67427b821e87802eb5a4ca4a5c038480e9a673f
|
| 3 |
+
size 15984
|
checkpoint-2000/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c967add958d82a4ef6285ec19e0e5a560f82b1ea3488260d2539d200cb5f199b
|
| 3 |
+
size 15984
|
checkpoint-2000/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1a5e30f320a5f3e4a4760af275c4a2bb9fdcfc6c661a3757da3b20b4c29a87fc
|
| 3 |
+
size 15984
|
checkpoint-2000/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3411b849d13fdd9e668c8cc5fafa8402de13fd4a2d5761eaf33f2791780721e4
|
| 3 |
+
size 15984
|
checkpoint-2000/rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1765109deada93bbc4f9e38b5c3c6bbd9afe9a12839bbbd72227ca7ff46467b1
|
| 3 |
+
size 15984
|
checkpoint-2000/rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6147f79a889f2906a81eea5c06f72acf722f674546fd6be8432d3d70a04392a8
|
| 3 |
+
size 15984
|
checkpoint-2000/rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a9f5815a19b60dc1bb303c451b515a4ec523fa0faf1c87c69b03ad4032c4baa2
|
| 3 |
+
size 15984
|
checkpoint-2000/rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:395da160bbc37f8049700e0d82d501cbd1effc732719599c0cd143db7892910e
|
| 3 |
+
size 15984
|
checkpoint-2000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfe4d506eb79b2afcda21ae799081e7329f98f2075ad245741a51d11e56354e4
|
| 3 |
+
size 1064
|
checkpoint-2000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<unk>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
checkpoint-2000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<unk>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<s>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"46331": {
|
| 28 |
+
"content": "<|sep|>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"46332": {
|
| 36 |
+
"content": "<|endoftext|>",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"46333": {
|
| 44 |
+
"content": "<|acc|>",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"46334": {
|
| 52 |
+
"content": "<|rrn|>",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"46335": {
|
| 60 |
+
"content": "<|tel|>",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
}
|
| 67 |
+
},
|
| 68 |
+
"bos_token": "<s>",
|
| 69 |
+
"clean_up_tokenization_spaces": false,
|
| 70 |
+
"eos_token": "</s>",
|
| 71 |
+
"legacy": false,
|
| 72 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 73 |
+
"pad_token": "<unk>",
|
| 74 |
+
"sp_model_kwargs": {},
|
| 75 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 76 |
+
"unk_token": "<unk>",
|
| 77 |
+
"use_default_system_prompt": true
|
| 78 |
+
}
|
checkpoint-2000/trainer_state.json
ADDED
|
@@ -0,0 +1,1219 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.607843137254903,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1,
|
| 13 |
+
"learning_rate": 0.0003996,
|
| 14 |
+
"loss": 9.2736,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.2,
|
| 19 |
+
"learning_rate": 0.0003992,
|
| 20 |
+
"loss": 7.7626,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.29,
|
| 25 |
+
"learning_rate": 0.00039880000000000004,
|
| 26 |
+
"loss": 7.5202,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.39,
|
| 31 |
+
"learning_rate": 0.00039840000000000003,
|
| 32 |
+
"loss": 7.371,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.49,
|
| 37 |
+
"learning_rate": 0.000398,
|
| 38 |
+
"loss": 7.2352,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.59,
|
| 43 |
+
"learning_rate": 0.0003976,
|
| 44 |
+
"loss": 7.1352,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.69,
|
| 49 |
+
"learning_rate": 0.0003972,
|
| 50 |
+
"loss": 7.0625,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.78,
|
| 55 |
+
"learning_rate": 0.0003968,
|
| 56 |
+
"loss": 6.911,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.88,
|
| 61 |
+
"learning_rate": 0.00039640000000000004,
|
| 62 |
+
"loss": 6.7717,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.98,
|
| 67 |
+
"learning_rate": 0.00039600000000000003,
|
| 68 |
+
"loss": 6.6484,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.08,
|
| 73 |
+
"learning_rate": 0.0003956,
|
| 74 |
+
"loss": 6.5492,
|
| 75 |
+
"step": 110
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 1.18,
|
| 79 |
+
"learning_rate": 0.0003952,
|
| 80 |
+
"loss": 6.4417,
|
| 81 |
+
"step": 120
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.27,
|
| 85 |
+
"learning_rate": 0.0003948,
|
| 86 |
+
"loss": 6.3311,
|
| 87 |
+
"step": 130
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.37,
|
| 91 |
+
"learning_rate": 0.0003944,
|
| 92 |
+
"loss": 6.2206,
|
| 93 |
+
"step": 140
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.47,
|
| 97 |
+
"learning_rate": 0.00039400000000000004,
|
| 98 |
+
"loss": 6.1179,
|
| 99 |
+
"step": 150
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.57,
|
| 103 |
+
"learning_rate": 0.0003936,
|
| 104 |
+
"loss": 6.0151,
|
| 105 |
+
"step": 160
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.67,
|
| 109 |
+
"learning_rate": 0.0003932,
|
| 110 |
+
"loss": 5.9035,
|
| 111 |
+
"step": 170
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.76,
|
| 115 |
+
"learning_rate": 0.0003928,
|
| 116 |
+
"loss": 5.8222,
|
| 117 |
+
"step": 180
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.86,
|
| 121 |
+
"learning_rate": 0.0003924,
|
| 122 |
+
"loss": 5.7042,
|
| 123 |
+
"step": 190
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.96,
|
| 127 |
+
"learning_rate": 0.000392,
|
| 128 |
+
"loss": 5.6265,
|
| 129 |
+
"step": 200
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.06,
|
| 133 |
+
"learning_rate": 0.00039160000000000003,
|
| 134 |
+
"loss": 5.5338,
|
| 135 |
+
"step": 210
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 2.16,
|
| 139 |
+
"learning_rate": 0.0003912,
|
| 140 |
+
"loss": 5.4521,
|
| 141 |
+
"step": 220
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 2.25,
|
| 145 |
+
"learning_rate": 0.0003908,
|
| 146 |
+
"loss": 5.3552,
|
| 147 |
+
"step": 230
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.35,
|
| 151 |
+
"learning_rate": 0.0003904,
|
| 152 |
+
"loss": 5.2771,
|
| 153 |
+
"step": 240
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.45,
|
| 157 |
+
"learning_rate": 0.00039,
|
| 158 |
+
"loss": 5.1587,
|
| 159 |
+
"step": 250
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.55,
|
| 163 |
+
"learning_rate": 0.0003896,
|
| 164 |
+
"loss": 5.0899,
|
| 165 |
+
"step": 260
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.65,
|
| 169 |
+
"learning_rate": 0.00038920000000000003,
|
| 170 |
+
"loss": 5.0191,
|
| 171 |
+
"step": 270
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.75,
|
| 175 |
+
"learning_rate": 0.0003888,
|
| 176 |
+
"loss": 4.9602,
|
| 177 |
+
"step": 280
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 2.84,
|
| 181 |
+
"learning_rate": 0.0003884,
|
| 182 |
+
"loss": 4.8366,
|
| 183 |
+
"step": 290
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 2.94,
|
| 187 |
+
"learning_rate": 0.000388,
|
| 188 |
+
"loss": 4.7848,
|
| 189 |
+
"step": 300
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 3.04,
|
| 193 |
+
"learning_rate": 0.0003876,
|
| 194 |
+
"loss": 4.7199,
|
| 195 |
+
"step": 310
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 3.14,
|
| 199 |
+
"learning_rate": 0.00038720000000000003,
|
| 200 |
+
"loss": 4.6306,
|
| 201 |
+
"step": 320
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 3.24,
|
| 205 |
+
"learning_rate": 0.0003868,
|
| 206 |
+
"loss": 4.5337,
|
| 207 |
+
"step": 330
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 3.33,
|
| 211 |
+
"learning_rate": 0.0003864,
|
| 212 |
+
"loss": 4.4796,
|
| 213 |
+
"step": 340
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 3.43,
|
| 217 |
+
"learning_rate": 0.000386,
|
| 218 |
+
"loss": 4.3881,
|
| 219 |
+
"step": 350
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 3.53,
|
| 223 |
+
"learning_rate": 0.0003856,
|
| 224 |
+
"loss": 4.2989,
|
| 225 |
+
"step": 360
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 3.63,
|
| 229 |
+
"learning_rate": 0.0003852,
|
| 230 |
+
"loss": 4.2533,
|
| 231 |
+
"step": 370
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 3.73,
|
| 235 |
+
"learning_rate": 0.00038480000000000003,
|
| 236 |
+
"loss": 4.2379,
|
| 237 |
+
"step": 380
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 3.82,
|
| 241 |
+
"learning_rate": 0.0003844,
|
| 242 |
+
"loss": 4.142,
|
| 243 |
+
"step": 390
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 3.92,
|
| 247 |
+
"learning_rate": 0.000384,
|
| 248 |
+
"loss": 4.0793,
|
| 249 |
+
"step": 400
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 4.02,
|
| 253 |
+
"learning_rate": 0.0003836,
|
| 254 |
+
"loss": 4.0005,
|
| 255 |
+
"step": 410
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 4.12,
|
| 259 |
+
"learning_rate": 0.0003832,
|
| 260 |
+
"loss": 3.9619,
|
| 261 |
+
"step": 420
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 4.22,
|
| 265 |
+
"learning_rate": 0.0003828,
|
| 266 |
+
"loss": 3.8687,
|
| 267 |
+
"step": 430
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 4.31,
|
| 271 |
+
"learning_rate": 0.0003824,
|
| 272 |
+
"loss": 3.8486,
|
| 273 |
+
"step": 440
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 4.41,
|
| 277 |
+
"learning_rate": 0.000382,
|
| 278 |
+
"loss": 3.7684,
|
| 279 |
+
"step": 450
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 4.51,
|
| 283 |
+
"learning_rate": 0.0003816,
|
| 284 |
+
"loss": 3.7013,
|
| 285 |
+
"step": 460
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 4.61,
|
| 289 |
+
"learning_rate": 0.0003812,
|
| 290 |
+
"loss": 3.6851,
|
| 291 |
+
"step": 470
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 4.71,
|
| 295 |
+
"learning_rate": 0.0003808,
|
| 296 |
+
"loss": 3.6585,
|
| 297 |
+
"step": 480
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.8,
|
| 301 |
+
"learning_rate": 0.0003804,
|
| 302 |
+
"loss": 3.6172,
|
| 303 |
+
"step": 490
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 4.9,
|
| 307 |
+
"learning_rate": 0.00038,
|
| 308 |
+
"loss": 3.5557,
|
| 309 |
+
"step": 500
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 5.0,
|
| 313 |
+
"learning_rate": 0.0003796,
|
| 314 |
+
"loss": 3.4746,
|
| 315 |
+
"step": 510
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 5.1,
|
| 319 |
+
"learning_rate": 0.0003792,
|
| 320 |
+
"loss": 3.4473,
|
| 321 |
+
"step": 520
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 5.2,
|
| 325 |
+
"learning_rate": 0.0003788,
|
| 326 |
+
"loss": 3.3828,
|
| 327 |
+
"step": 530
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 5.29,
|
| 331 |
+
"learning_rate": 0.0003784,
|
| 332 |
+
"loss": 3.3868,
|
| 333 |
+
"step": 540
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 5.39,
|
| 337 |
+
"learning_rate": 0.00037799999999999997,
|
| 338 |
+
"loss": 3.31,
|
| 339 |
+
"step": 550
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 5.49,
|
| 343 |
+
"learning_rate": 0.0003776,
|
| 344 |
+
"loss": 3.2628,
|
| 345 |
+
"step": 560
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 5.59,
|
| 349 |
+
"learning_rate": 0.0003772,
|
| 350 |
+
"loss": 3.2541,
|
| 351 |
+
"step": 570
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 5.69,
|
| 355 |
+
"learning_rate": 0.0003768,
|
| 356 |
+
"loss": 3.223,
|
| 357 |
+
"step": 580
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 5.78,
|
| 361 |
+
"learning_rate": 0.0003764,
|
| 362 |
+
"loss": 3.2028,
|
| 363 |
+
"step": 590
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 5.88,
|
| 367 |
+
"learning_rate": 0.000376,
|
| 368 |
+
"loss": 3.1659,
|
| 369 |
+
"step": 600
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 5.98,
|
| 373 |
+
"learning_rate": 0.0003756,
|
| 374 |
+
"loss": 3.0847,
|
| 375 |
+
"step": 610
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 6.08,
|
| 379 |
+
"learning_rate": 0.0003752,
|
| 380 |
+
"loss": 3.0215,
|
| 381 |
+
"step": 620
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 6.18,
|
| 385 |
+
"learning_rate": 0.00037480000000000006,
|
| 386 |
+
"loss": 3.0149,
|
| 387 |
+
"step": 630
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 6.27,
|
| 391 |
+
"learning_rate": 0.00037440000000000005,
|
| 392 |
+
"loss": 3.0177,
|
| 393 |
+
"step": 640
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 6.37,
|
| 397 |
+
"learning_rate": 0.00037400000000000004,
|
| 398 |
+
"loss": 2.9474,
|
| 399 |
+
"step": 650
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 6.47,
|
| 403 |
+
"learning_rate": 0.00037360000000000003,
|
| 404 |
+
"loss": 2.9245,
|
| 405 |
+
"step": 660
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 6.57,
|
| 409 |
+
"learning_rate": 0.0003732,
|
| 410 |
+
"loss": 2.9218,
|
| 411 |
+
"step": 670
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 6.67,
|
| 415 |
+
"learning_rate": 0.00037280000000000006,
|
| 416 |
+
"loss": 2.8666,
|
| 417 |
+
"step": 680
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 6.76,
|
| 421 |
+
"learning_rate": 0.00037240000000000005,
|
| 422 |
+
"loss": 2.8821,
|
| 423 |
+
"step": 690
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 6.86,
|
| 427 |
+
"learning_rate": 0.00037200000000000004,
|
| 428 |
+
"loss": 2.8243,
|
| 429 |
+
"step": 700
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 6.96,
|
| 433 |
+
"learning_rate": 0.00037160000000000003,
|
| 434 |
+
"loss": 2.7753,
|
| 435 |
+
"step": 710
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 7.06,
|
| 439 |
+
"learning_rate": 0.0003712,
|
| 440 |
+
"loss": 2.7086,
|
| 441 |
+
"step": 720
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 7.16,
|
| 445 |
+
"learning_rate": 0.0003708,
|
| 446 |
+
"loss": 2.7104,
|
| 447 |
+
"step": 730
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 7.25,
|
| 451 |
+
"learning_rate": 0.00037040000000000006,
|
| 452 |
+
"loss": 2.7103,
|
| 453 |
+
"step": 740
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 7.35,
|
| 457 |
+
"learning_rate": 0.00037000000000000005,
|
| 458 |
+
"loss": 2.6707,
|
| 459 |
+
"step": 750
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 7.45,
|
| 463 |
+
"learning_rate": 0.00036960000000000004,
|
| 464 |
+
"loss": 2.6413,
|
| 465 |
+
"step": 760
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 7.55,
|
| 469 |
+
"learning_rate": 0.00036920000000000003,
|
| 470 |
+
"loss": 2.6359,
|
| 471 |
+
"step": 770
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 7.65,
|
| 475 |
+
"learning_rate": 0.0003688,
|
| 476 |
+
"loss": 2.5838,
|
| 477 |
+
"step": 780
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 7.75,
|
| 481 |
+
"learning_rate": 0.0003684,
|
| 482 |
+
"loss": 2.6212,
|
| 483 |
+
"step": 790
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 7.84,
|
| 487 |
+
"learning_rate": 0.00036800000000000005,
|
| 488 |
+
"loss": 2.5718,
|
| 489 |
+
"step": 800
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 7.94,
|
| 493 |
+
"learning_rate": 0.00036760000000000004,
|
| 494 |
+
"loss": 2.5348,
|
| 495 |
+
"step": 810
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 8.04,
|
| 499 |
+
"learning_rate": 0.00036720000000000004,
|
| 500 |
+
"loss": 2.4195,
|
| 501 |
+
"step": 820
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 8.14,
|
| 505 |
+
"learning_rate": 0.0003668,
|
| 506 |
+
"loss": 2.4938,
|
| 507 |
+
"step": 830
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 8.24,
|
| 511 |
+
"learning_rate": 0.0003664,
|
| 512 |
+
"loss": 2.4372,
|
| 513 |
+
"step": 840
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 8.33,
|
| 517 |
+
"learning_rate": 0.000366,
|
| 518 |
+
"loss": 2.4567,
|
| 519 |
+
"step": 850
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 8.43,
|
| 523 |
+
"learning_rate": 0.00036560000000000005,
|
| 524 |
+
"loss": 2.4108,
|
| 525 |
+
"step": 860
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 8.53,
|
| 529 |
+
"learning_rate": 0.00036520000000000004,
|
| 530 |
+
"loss": 2.3993,
|
| 531 |
+
"step": 870
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 8.63,
|
| 535 |
+
"learning_rate": 0.00036480000000000003,
|
| 536 |
+
"loss": 2.3739,
|
| 537 |
+
"step": 880
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 8.73,
|
| 541 |
+
"learning_rate": 0.0003644,
|
| 542 |
+
"loss": 2.391,
|
| 543 |
+
"step": 890
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 8.82,
|
| 547 |
+
"learning_rate": 0.000364,
|
| 548 |
+
"loss": 2.38,
|
| 549 |
+
"step": 900
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 8.92,
|
| 553 |
+
"learning_rate": 0.00036360000000000006,
|
| 554 |
+
"loss": 2.3257,
|
| 555 |
+
"step": 910
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 9.02,
|
| 559 |
+
"learning_rate": 0.00036320000000000005,
|
| 560 |
+
"loss": 2.1956,
|
| 561 |
+
"step": 920
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 9.12,
|
| 565 |
+
"learning_rate": 0.00036280000000000004,
|
| 566 |
+
"loss": 2.2944,
|
| 567 |
+
"step": 930
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 9.22,
|
| 571 |
+
"learning_rate": 0.0003624,
|
| 572 |
+
"loss": 2.2163,
|
| 573 |
+
"step": 940
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 9.31,
|
| 577 |
+
"learning_rate": 0.000362,
|
| 578 |
+
"loss": 2.2641,
|
| 579 |
+
"step": 950
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 9.41,
|
| 583 |
+
"learning_rate": 0.0003616,
|
| 584 |
+
"loss": 2.2035,
|
| 585 |
+
"step": 960
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 9.51,
|
| 589 |
+
"learning_rate": 0.00036120000000000005,
|
| 590 |
+
"loss": 2.188,
|
| 591 |
+
"step": 970
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 9.61,
|
| 595 |
+
"learning_rate": 0.00036080000000000004,
|
| 596 |
+
"loss": 2.197,
|
| 597 |
+
"step": 980
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 9.71,
|
| 601 |
+
"learning_rate": 0.00036040000000000003,
|
| 602 |
+
"loss": 2.18,
|
| 603 |
+
"step": 990
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 9.8,
|
| 607 |
+
"learning_rate": 0.00036,
|
| 608 |
+
"loss": 2.1994,
|
| 609 |
+
"step": 1000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 9.9,
|
| 613 |
+
"learning_rate": 0.0003596,
|
| 614 |
+
"loss": 2.1538,
|
| 615 |
+
"step": 1010
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 10.0,
|
| 619 |
+
"learning_rate": 0.0003592,
|
| 620 |
+
"loss": 2.0495,
|
| 621 |
+
"step": 1020
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 10.1,
|
| 625 |
+
"learning_rate": 0.00035880000000000005,
|
| 626 |
+
"loss": 2.0501,
|
| 627 |
+
"step": 1030
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 10.2,
|
| 631 |
+
"learning_rate": 0.00035840000000000004,
|
| 632 |
+
"loss": 2.0302,
|
| 633 |
+
"step": 1040
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 10.29,
|
| 637 |
+
"learning_rate": 0.00035800000000000003,
|
| 638 |
+
"loss": 2.0996,
|
| 639 |
+
"step": 1050
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 10.39,
|
| 643 |
+
"learning_rate": 0.0003576,
|
| 644 |
+
"loss": 2.0369,
|
| 645 |
+
"step": 1060
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 10.49,
|
| 649 |
+
"learning_rate": 0.0003572,
|
| 650 |
+
"loss": 1.9996,
|
| 651 |
+
"step": 1070
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 10.59,
|
| 655 |
+
"learning_rate": 0.0003568,
|
| 656 |
+
"loss": 2.0254,
|
| 657 |
+
"step": 1080
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 10.69,
|
| 661 |
+
"learning_rate": 0.00035640000000000004,
|
| 662 |
+
"loss": 2.0004,
|
| 663 |
+
"step": 1090
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 10.78,
|
| 667 |
+
"learning_rate": 0.00035600000000000003,
|
| 668 |
+
"loss": 2.0154,
|
| 669 |
+
"step": 1100
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 10.88,
|
| 673 |
+
"learning_rate": 0.0003556,
|
| 674 |
+
"loss": 2.0014,
|
| 675 |
+
"step": 1110
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 10.98,
|
| 679 |
+
"learning_rate": 0.00035524,
|
| 680 |
+
"loss": 1.9377,
|
| 681 |
+
"step": 1120
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 11.08,
|
| 685 |
+
"learning_rate": 0.00035484000000000004,
|
| 686 |
+
"loss": 1.8296,
|
| 687 |
+
"step": 1130
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 11.18,
|
| 691 |
+
"learning_rate": 0.00035444000000000003,
|
| 692 |
+
"loss": 1.9002,
|
| 693 |
+
"step": 1140
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 11.27,
|
| 697 |
+
"learning_rate": 0.00035404,
|
| 698 |
+
"loss": 1.9336,
|
| 699 |
+
"step": 1150
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 11.37,
|
| 703 |
+
"learning_rate": 0.00035364,
|
| 704 |
+
"loss": 1.9041,
|
| 705 |
+
"step": 1160
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 11.47,
|
| 709 |
+
"learning_rate": 0.00035324,
|
| 710 |
+
"loss": 1.8324,
|
| 711 |
+
"step": 1170
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 11.57,
|
| 715 |
+
"learning_rate": 0.00035284,
|
| 716 |
+
"loss": 1.8663,
|
| 717 |
+
"step": 1180
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 11.67,
|
| 721 |
+
"learning_rate": 0.00035244000000000003,
|
| 722 |
+
"loss": 1.8358,
|
| 723 |
+
"step": 1190
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 11.76,
|
| 727 |
+
"learning_rate": 0.00035204,
|
| 728 |
+
"loss": 1.8586,
|
| 729 |
+
"step": 1200
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 11.86,
|
| 733 |
+
"learning_rate": 0.00035164,
|
| 734 |
+
"loss": 1.8452,
|
| 735 |
+
"step": 1210
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 11.96,
|
| 739 |
+
"learning_rate": 0.00035124,
|
| 740 |
+
"loss": 1.7957,
|
| 741 |
+
"step": 1220
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 12.06,
|
| 745 |
+
"learning_rate": 0.00035084,
|
| 746 |
+
"loss": 1.6603,
|
| 747 |
+
"step": 1230
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 12.16,
|
| 751 |
+
"learning_rate": 0.00035044000000000004,
|
| 752 |
+
"loss": 1.7903,
|
| 753 |
+
"step": 1240
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 12.25,
|
| 757 |
+
"learning_rate": 0.00035004000000000003,
|
| 758 |
+
"loss": 1.7574,
|
| 759 |
+
"step": 1250
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 12.35,
|
| 763 |
+
"learning_rate": 0.00034964,
|
| 764 |
+
"loss": 1.7823,
|
| 765 |
+
"step": 1260
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 12.45,
|
| 769 |
+
"learning_rate": 0.00034924,
|
| 770 |
+
"loss": 1.7038,
|
| 771 |
+
"step": 1270
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 12.55,
|
| 775 |
+
"learning_rate": 0.00034884,
|
| 776 |
+
"loss": 1.7186,
|
| 777 |
+
"step": 1280
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 12.65,
|
| 781 |
+
"learning_rate": 0.00034844,
|
| 782 |
+
"loss": 1.6948,
|
| 783 |
+
"step": 1290
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 12.75,
|
| 787 |
+
"learning_rate": 0.00034804000000000004,
|
| 788 |
+
"loss": 1.7347,
|
| 789 |
+
"step": 1300
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 12.84,
|
| 793 |
+
"learning_rate": 0.00034764,
|
| 794 |
+
"loss": 1.7304,
|
| 795 |
+
"step": 1310
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 12.94,
|
| 799 |
+
"learning_rate": 0.00034724,
|
| 800 |
+
"loss": 1.6862,
|
| 801 |
+
"step": 1320
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 13.04,
|
| 805 |
+
"learning_rate": 0.00034684,
|
| 806 |
+
"loss": 1.4857,
|
| 807 |
+
"step": 1330
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 13.14,
|
| 811 |
+
"learning_rate": 0.00034644,
|
| 812 |
+
"loss": 1.6981,
|
| 813 |
+
"step": 1340
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 13.24,
|
| 817 |
+
"learning_rate": 0.00034604,
|
| 818 |
+
"loss": 1.5926,
|
| 819 |
+
"step": 1350
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 13.33,
|
| 823 |
+
"learning_rate": 0.00034564000000000003,
|
| 824 |
+
"loss": 1.6545,
|
| 825 |
+
"step": 1360
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 13.43,
|
| 829 |
+
"learning_rate": 0.00034524,
|
| 830 |
+
"loss": 1.5918,
|
| 831 |
+
"step": 1370
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 13.53,
|
| 835 |
+
"learning_rate": 0.00034484,
|
| 836 |
+
"loss": 1.5743,
|
| 837 |
+
"step": 1380
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 13.63,
|
| 841 |
+
"learning_rate": 0.00034444,
|
| 842 |
+
"loss": 1.5867,
|
| 843 |
+
"step": 1390
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 13.73,
|
| 847 |
+
"learning_rate": 0.00034404,
|
| 848 |
+
"loss": 1.6042,
|
| 849 |
+
"step": 1400
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 13.82,
|
| 853 |
+
"learning_rate": 0.00034364,
|
| 854 |
+
"loss": 1.6308,
|
| 855 |
+
"step": 1410
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 13.92,
|
| 859 |
+
"learning_rate": 0.00034324000000000003,
|
| 860 |
+
"loss": 1.5813,
|
| 861 |
+
"step": 1420
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 14.02,
|
| 865 |
+
"learning_rate": 0.00034284,
|
| 866 |
+
"loss": 1.364,
|
| 867 |
+
"step": 1430
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 14.12,
|
| 871 |
+
"learning_rate": 0.00034244,
|
| 872 |
+
"loss": 1.5849,
|
| 873 |
+
"step": 1440
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 14.22,
|
| 877 |
+
"learning_rate": 0.00034204,
|
| 878 |
+
"loss": 1.4957,
|
| 879 |
+
"step": 1450
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 14.31,
|
| 883 |
+
"learning_rate": 0.00034164,
|
| 884 |
+
"loss": 1.5312,
|
| 885 |
+
"step": 1460
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 14.41,
|
| 889 |
+
"learning_rate": 0.00034124,
|
| 890 |
+
"loss": 1.4995,
|
| 891 |
+
"step": 1470
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 14.51,
|
| 895 |
+
"learning_rate": 0.00034084,
|
| 896 |
+
"loss": 1.4501,
|
| 897 |
+
"step": 1480
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 14.61,
|
| 901 |
+
"learning_rate": 0.00034044,
|
| 902 |
+
"loss": 1.4721,
|
| 903 |
+
"step": 1490
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 14.71,
|
| 907 |
+
"learning_rate": 0.00034004,
|
| 908 |
+
"loss": 1.4792,
|
| 909 |
+
"step": 1500
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 14.8,
|
| 913 |
+
"learning_rate": 0.00033964,
|
| 914 |
+
"loss": 1.513,
|
| 915 |
+
"step": 1510
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 14.9,
|
| 919 |
+
"learning_rate": 0.00033924,
|
| 920 |
+
"loss": 1.5028,
|
| 921 |
+
"step": 1520
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 15.0,
|
| 925 |
+
"learning_rate": 0.00033884000000000003,
|
| 926 |
+
"loss": 1.3511,
|
| 927 |
+
"step": 1530
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 15.1,
|
| 931 |
+
"learning_rate": 0.00033844,
|
| 932 |
+
"loss": 1.4152,
|
| 933 |
+
"step": 1540
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 15.2,
|
| 937 |
+
"learning_rate": 0.00033804,
|
| 938 |
+
"loss": 1.4216,
|
| 939 |
+
"step": 1550
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 15.29,
|
| 943 |
+
"learning_rate": 0.00033764,
|
| 944 |
+
"loss": 1.4177,
|
| 945 |
+
"step": 1560
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 15.39,
|
| 949 |
+
"learning_rate": 0.00033724,
|
| 950 |
+
"loss": 1.4074,
|
| 951 |
+
"step": 1570
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 15.49,
|
| 955 |
+
"learning_rate": 0.00033684,
|
| 956 |
+
"loss": 1.3532,
|
| 957 |
+
"step": 1580
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 15.59,
|
| 961 |
+
"learning_rate": 0.00033644,
|
| 962 |
+
"loss": 1.3525,
|
| 963 |
+
"step": 1590
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 15.69,
|
| 967 |
+
"learning_rate": 0.00033604,
|
| 968 |
+
"loss": 1.3755,
|
| 969 |
+
"step": 1600
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 15.78,
|
| 973 |
+
"learning_rate": 0.00033564,
|
| 974 |
+
"loss": 1.3989,
|
| 975 |
+
"step": 1610
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 15.88,
|
| 979 |
+
"learning_rate": 0.00033524,
|
| 980 |
+
"loss": 1.4012,
|
| 981 |
+
"step": 1620
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 15.98,
|
| 985 |
+
"learning_rate": 0.00033484,
|
| 986 |
+
"loss": 1.3386,
|
| 987 |
+
"step": 1630
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 16.08,
|
| 991 |
+
"learning_rate": 0.00033444,
|
| 992 |
+
"loss": 1.2476,
|
| 993 |
+
"step": 1640
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 16.18,
|
| 997 |
+
"learning_rate": 0.00033404,
|
| 998 |
+
"loss": 1.3435,
|
| 999 |
+
"step": 1650
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 16.27,
|
| 1003 |
+
"learning_rate": 0.00033364,
|
| 1004 |
+
"loss": 1.3131,
|
| 1005 |
+
"step": 1660
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 16.37,
|
| 1009 |
+
"learning_rate": 0.00033324,
|
| 1010 |
+
"loss": 1.3099,
|
| 1011 |
+
"step": 1670
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 16.47,
|
| 1015 |
+
"learning_rate": 0.00033284,
|
| 1016 |
+
"loss": 1.2693,
|
| 1017 |
+
"step": 1680
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 16.57,
|
| 1021 |
+
"learning_rate": 0.00033244,
|
| 1022 |
+
"loss": 1.2649,
|
| 1023 |
+
"step": 1690
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 16.67,
|
| 1027 |
+
"learning_rate": 0.00033203999999999997,
|
| 1028 |
+
"loss": 1.2777,
|
| 1029 |
+
"step": 1700
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 16.76,
|
| 1033 |
+
"learning_rate": 0.00033164,
|
| 1034 |
+
"loss": 1.2889,
|
| 1035 |
+
"step": 1710
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 16.86,
|
| 1039 |
+
"learning_rate": 0.00033124,
|
| 1040 |
+
"loss": 1.2729,
|
| 1041 |
+
"step": 1720
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 16.96,
|
| 1045 |
+
"learning_rate": 0.00033084,
|
| 1046 |
+
"loss": 1.2529,
|
| 1047 |
+
"step": 1730
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 17.06,
|
| 1051 |
+
"learning_rate": 0.00033044,
|
| 1052 |
+
"loss": 1.1566,
|
| 1053 |
+
"step": 1740
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 17.16,
|
| 1057 |
+
"learning_rate": 0.00033004,
|
| 1058 |
+
"loss": 1.2668,
|
| 1059 |
+
"step": 1750
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 17.25,
|
| 1063 |
+
"learning_rate": 0.00032964,
|
| 1064 |
+
"loss": 1.2052,
|
| 1065 |
+
"step": 1760
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 17.35,
|
| 1069 |
+
"learning_rate": 0.00032924,
|
| 1070 |
+
"loss": 1.2159,
|
| 1071 |
+
"step": 1770
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 17.45,
|
| 1075 |
+
"learning_rate": 0.00032884000000000006,
|
| 1076 |
+
"loss": 1.2003,
|
| 1077 |
+
"step": 1780
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 17.55,
|
| 1081 |
+
"learning_rate": 0.00032844000000000005,
|
| 1082 |
+
"loss": 1.1836,
|
| 1083 |
+
"step": 1790
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 17.65,
|
| 1087 |
+
"learning_rate": 0.00032804000000000004,
|
| 1088 |
+
"loss": 1.1857,
|
| 1089 |
+
"step": 1800
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 17.75,
|
| 1093 |
+
"learning_rate": 0.00032764000000000003,
|
| 1094 |
+
"loss": 1.2012,
|
| 1095 |
+
"step": 1810
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 17.84,
|
| 1099 |
+
"learning_rate": 0.00032724,
|
| 1100 |
+
"loss": 1.1907,
|
| 1101 |
+
"step": 1820
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 17.94,
|
| 1105 |
+
"learning_rate": 0.00032684,
|
| 1106 |
+
"loss": 1.1723,
|
| 1107 |
+
"step": 1830
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 18.04,
|
| 1111 |
+
"learning_rate": 0.00032644000000000005,
|
| 1112 |
+
"loss": 1.0299,
|
| 1113 |
+
"step": 1840
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 18.14,
|
| 1117 |
+
"learning_rate": 0.00032604000000000004,
|
| 1118 |
+
"loss": 1.1996,
|
| 1119 |
+
"step": 1850
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 18.24,
|
| 1123 |
+
"learning_rate": 0.00032564000000000003,
|
| 1124 |
+
"loss": 1.0919,
|
| 1125 |
+
"step": 1860
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 18.33,
|
| 1129 |
+
"learning_rate": 0.00032524,
|
| 1130 |
+
"loss": 1.1291,
|
| 1131 |
+
"step": 1870
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 18.43,
|
| 1135 |
+
"learning_rate": 0.00032484,
|
| 1136 |
+
"loss": 1.1343,
|
| 1137 |
+
"step": 1880
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 18.53,
|
| 1141 |
+
"learning_rate": 0.00032444000000000006,
|
| 1142 |
+
"loss": 1.0965,
|
| 1143 |
+
"step": 1890
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 18.63,
|
| 1147 |
+
"learning_rate": 0.00032404000000000005,
|
| 1148 |
+
"loss": 1.1149,
|
| 1149 |
+
"step": 1900
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 18.73,
|
| 1153 |
+
"learning_rate": 0.00032364000000000004,
|
| 1154 |
+
"loss": 1.132,
|
| 1155 |
+
"step": 1910
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 18.82,
|
| 1159 |
+
"learning_rate": 0.00032324000000000003,
|
| 1160 |
+
"loss": 1.1218,
|
| 1161 |
+
"step": 1920
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 18.92,
|
| 1165 |
+
"learning_rate": 0.00032284,
|
| 1166 |
+
"loss": 1.0928,
|
| 1167 |
+
"step": 1930
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 19.02,
|
| 1171 |
+
"learning_rate": 0.00032244,
|
| 1172 |
+
"loss": 0.9377,
|
| 1173 |
+
"step": 1940
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 19.12,
|
| 1177 |
+
"learning_rate": 0.00032204000000000005,
|
| 1178 |
+
"loss": 1.1132,
|
| 1179 |
+
"step": 1950
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 19.22,
|
| 1183 |
+
"learning_rate": 0.00032164000000000004,
|
| 1184 |
+
"loss": 1.0275,
|
| 1185 |
+
"step": 1960
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 19.31,
|
| 1189 |
+
"learning_rate": 0.00032124000000000003,
|
| 1190 |
+
"loss": 1.0312,
|
| 1191 |
+
"step": 1970
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 19.41,
|
| 1195 |
+
"learning_rate": 0.00032084,
|
| 1196 |
+
"loss": 1.0248,
|
| 1197 |
+
"step": 1980
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 19.51,
|
| 1201 |
+
"learning_rate": 0.00032044,
|
| 1202 |
+
"loss": 1.0063,
|
| 1203 |
+
"step": 1990
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 19.61,
|
| 1207 |
+
"learning_rate": 0.00032004,
|
| 1208 |
+
"loss": 1.0438,
|
| 1209 |
+
"step": 2000
|
| 1210 |
+
}
|
| 1211 |
+
],
|
| 1212 |
+
"logging_steps": 10,
|
| 1213 |
+
"max_steps": 10000,
|
| 1214 |
+
"num_train_epochs": 99,
|
| 1215 |
+
"save_steps": 1000,
|
| 1216 |
+
"total_flos": 1.33620861763584e+17,
|
| 1217 |
+
"trial_name": null,
|
| 1218 |
+
"trial_params": null
|
| 1219 |
+
}
|
checkpoint-2000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:36b5d44c73b8e982d1bd71774d1ff8b2323fbcd973d1b01903f213cf13c78cc7
|
| 3 |
+
size 5880
|