Training in progress, step 20000, checkpoint
Browse files- checkpoint-20000/config.json +32 -0
- checkpoint-20000/generation_config.json +6 -0
- checkpoint-20000/model.safetensors +3 -0
- checkpoint-20000/optimizer.pt +3 -0
- checkpoint-20000/rng_state.pth +3 -0
- checkpoint-20000/scheduler.pt +3 -0
- checkpoint-20000/special_tokens_map.json +24 -0
- checkpoint-20000/tokenizer.json +96 -0
- checkpoint-20000/tokenizer_config.json +38 -0
- checkpoint-20000/trainer_state.json +1433 -0
- checkpoint-20000/training_args.bin +3 -0
checkpoint-20000/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "EleutherAI/pythia-14m",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPTNeoXForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": true,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"classifier_dropout": 0.1,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout": 0.0,
|
| 13 |
+
"hidden_size": 128,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 512,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 2048,
|
| 18 |
+
"model_type": "gpt_neox",
|
| 19 |
+
"num_attention_heads": 4,
|
| 20 |
+
"num_hidden_layers": 6,
|
| 21 |
+
"partial_rotary_factor": 0.25,
|
| 22 |
+
"rope_scaling": null,
|
| 23 |
+
"rope_theta": 10000,
|
| 24 |
+
"rotary_emb_base": 10000,
|
| 25 |
+
"rotary_pct": 0.25,
|
| 26 |
+
"tie_word_embeddings": false,
|
| 27 |
+
"torch_dtype": "float32",
|
| 28 |
+
"transformers_version": "4.45.2",
|
| 29 |
+
"use_cache": true,
|
| 30 |
+
"use_parallel_residual": true,
|
| 31 |
+
"vocab_size": 8
|
| 32 |
+
}
|
checkpoint-20000/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"transformers_version": "4.45.2"
|
| 6 |
+
}
|
checkpoint-20000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c9b770647b28acef8afdbfd9e7c154beccce0142bd85c458606657188104820b
|
| 3 |
+
size 4776024
|
checkpoint-20000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b816320cebb45156c7890b5263b67b4a5a3c3e2f8853f0b88cc8d6315c44c7e
|
| 3 |
+
size 9598906
|
checkpoint-20000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d9cd6a0487226e5bd30d1846894c82af483733ab4381b75bae9c0745e05d405
|
| 3 |
+
size 14244
|
checkpoint-20000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:11994d962713fb3ecf43b552d5afdf46c657cbafe99985ebaa731400e961d792
|
| 3 |
+
size 1000
|
checkpoint-20000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|beginoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "<|endoftext|>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<|unknown|>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
checkpoint-20000/tokenizer.json
ADDED
|
@@ -0,0 +1,96 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"version": "1.0",
|
| 3 |
+
"truncation": null,
|
| 4 |
+
"padding": null,
|
| 5 |
+
"added_tokens": [
|
| 6 |
+
{
|
| 7 |
+
"id": 0,
|
| 8 |
+
"content": "<|beginoftext|>",
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"lstrip": false,
|
| 11 |
+
"rstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"special": true
|
| 14 |
+
},
|
| 15 |
+
{
|
| 16 |
+
"id": 1,
|
| 17 |
+
"content": "<|endoftext|>",
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"normalized": false,
|
| 22 |
+
"special": true
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"id": 2,
|
| 26 |
+
"content": "<|unknown|>",
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"rstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"special": true
|
| 32 |
+
}
|
| 33 |
+
],
|
| 34 |
+
"normalizer": {
|
| 35 |
+
"type": "NFC"
|
| 36 |
+
},
|
| 37 |
+
"pre_tokenizer": {
|
| 38 |
+
"type": "ByteLevel",
|
| 39 |
+
"add_prefix_space": false,
|
| 40 |
+
"trim_offsets": true,
|
| 41 |
+
"use_regex": true
|
| 42 |
+
},
|
| 43 |
+
"post_processor": {
|
| 44 |
+
"type": "TemplateProcessing",
|
| 45 |
+
"single": [
|
| 46 |
+
{
|
| 47 |
+
"Sequence": {
|
| 48 |
+
"id": "A",
|
| 49 |
+
"type_id": 0
|
| 50 |
+
}
|
| 51 |
+
}
|
| 52 |
+
],
|
| 53 |
+
"pair": [
|
| 54 |
+
{
|
| 55 |
+
"Sequence": {
|
| 56 |
+
"id": "A",
|
| 57 |
+
"type_id": 0
|
| 58 |
+
}
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"Sequence": {
|
| 62 |
+
"id": "B",
|
| 63 |
+
"type_id": 1
|
| 64 |
+
}
|
| 65 |
+
}
|
| 66 |
+
],
|
| 67 |
+
"special_tokens": {}
|
| 68 |
+
},
|
| 69 |
+
"decoder": {
|
| 70 |
+
"type": "ByteLevel",
|
| 71 |
+
"add_prefix_space": false,
|
| 72 |
+
"trim_offsets": true,
|
| 73 |
+
"use_regex": true
|
| 74 |
+
},
|
| 75 |
+
"model": {
|
| 76 |
+
"type": "BPE",
|
| 77 |
+
"dropout": null,
|
| 78 |
+
"unk_token": "<|unknown|>",
|
| 79 |
+
"continuing_subword_prefix": null,
|
| 80 |
+
"end_of_word_suffix": null,
|
| 81 |
+
"fuse_unk": false,
|
| 82 |
+
"byte_fallback": false,
|
| 83 |
+
"ignore_merges": false,
|
| 84 |
+
"vocab": {
|
| 85 |
+
"<|beginoftext|>": 0,
|
| 86 |
+
"<|padding|>": 1,
|
| 87 |
+
"<|unknown|>": 2,
|
| 88 |
+
"a": 3,
|
| 89 |
+
"c": 4,
|
| 90 |
+
"g": 5,
|
| 91 |
+
"t": 6,
|
| 92 |
+
"n": 7
|
| 93 |
+
},
|
| 94 |
+
"merges": []
|
| 95 |
+
}
|
| 96 |
+
}
|
checkpoint-20000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": false,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<|beginoftext|>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": true,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<|endoftext|>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "<|unknown|>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": "<|beginoftext|>",
|
| 32 |
+
"clean_up_tokenization_spaces": true,
|
| 33 |
+
"eos_token": "<|endoftext|>",
|
| 34 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 35 |
+
"pad_token": "<|endoftext|>",
|
| 36 |
+
"tokenizer_class": "GPTNeoXTokenizer",
|
| 37 |
+
"unk_token": "<|unknown|>"
|
| 38 |
+
}
|
checkpoint-20000/trainer_state.json
ADDED
|
@@ -0,0 +1,1433 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7817385866166354,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 20000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.003908692933083177,
|
| 13 |
+
"grad_norm": 10707.548828125,
|
| 14 |
+
"learning_rate": 4.98455462579182e-05,
|
| 15 |
+
"loss": 2311.9475,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.007817385866166354,
|
| 20 |
+
"grad_norm": 7085.59814453125,
|
| 21 |
+
"learning_rate": 4.9650035191991866e-05,
|
| 22 |
+
"loss": 2032.1091,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01172607879924953,
|
| 27 |
+
"grad_norm": 7845.0673828125,
|
| 28 |
+
"learning_rate": 4.945452412606554e-05,
|
| 29 |
+
"loss": 1999.2905,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.015634771732332707,
|
| 34 |
+
"grad_norm": 7440.90087890625,
|
| 35 |
+
"learning_rate": 4.925901306013921e-05,
|
| 36 |
+
"loss": 2000.142,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.019543464665415886,
|
| 41 |
+
"grad_norm": 8261.16796875,
|
| 42 |
+
"learning_rate": 4.9063501994212875e-05,
|
| 43 |
+
"loss": 1979.9078,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02345215759849906,
|
| 48 |
+
"grad_norm": 10481.5185546875,
|
| 49 |
+
"learning_rate": 4.886799092828655e-05,
|
| 50 |
+
"loss": 1976.0278,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02736085053158224,
|
| 55 |
+
"grad_norm": 4640.42236328125,
|
| 56 |
+
"learning_rate": 4.867247986236021e-05,
|
| 57 |
+
"loss": 1975.9683,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.031269543464665414,
|
| 62 |
+
"grad_norm": 9219.927734375,
|
| 63 |
+
"learning_rate": 4.847696879643388e-05,
|
| 64 |
+
"loss": 1966.8808,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03517823639774859,
|
| 69 |
+
"grad_norm": 9850.56640625,
|
| 70 |
+
"learning_rate": 4.828145773050755e-05,
|
| 71 |
+
"loss": 1963.0811,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03908692933083177,
|
| 76 |
+
"grad_norm": 11089.212890625,
|
| 77 |
+
"learning_rate": 4.8085946664581216e-05,
|
| 78 |
+
"loss": 1962.5177,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04299562226391495,
|
| 83 |
+
"grad_norm": 4673.15087890625,
|
| 84 |
+
"learning_rate": 4.789043559865489e-05,
|
| 85 |
+
"loss": 1957.4523,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04690431519699812,
|
| 90 |
+
"grad_norm": 4652.90771484375,
|
| 91 |
+
"learning_rate": 4.7694924532728554e-05,
|
| 92 |
+
"loss": 1959.5081,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.0508130081300813,
|
| 97 |
+
"grad_norm": 9705.9951171875,
|
| 98 |
+
"learning_rate": 4.7499413466802226e-05,
|
| 99 |
+
"loss": 1960.9455,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.05472170106316448,
|
| 104 |
+
"grad_norm": 5659.09375,
|
| 105 |
+
"learning_rate": 4.730390240087589e-05,
|
| 106 |
+
"loss": 1954.4797,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05863039399624766,
|
| 111 |
+
"grad_norm": 6702.53857421875,
|
| 112 |
+
"learning_rate": 4.7108391334949557e-05,
|
| 113 |
+
"loss": 1953.1802,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.06253908692933083,
|
| 118 |
+
"grad_norm": 9205.0869140625,
|
| 119 |
+
"learning_rate": 4.691288026902323e-05,
|
| 120 |
+
"loss": 1953.72,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06644777986241401,
|
| 125 |
+
"grad_norm": 9152.9912109375,
|
| 126 |
+
"learning_rate": 4.67173692030969e-05,
|
| 127 |
+
"loss": 1954.9048,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.07035647279549719,
|
| 132 |
+
"grad_norm": 6856.97705078125,
|
| 133 |
+
"learning_rate": 4.6521858137170566e-05,
|
| 134 |
+
"loss": 1948.0663,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07426516572858036,
|
| 139 |
+
"grad_norm": 10148.6259765625,
|
| 140 |
+
"learning_rate": 4.632634707124424e-05,
|
| 141 |
+
"loss": 1947.7772,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07817385866166354,
|
| 146 |
+
"grad_norm": 7209.80908203125,
|
| 147 |
+
"learning_rate": 4.6130836005317904e-05,
|
| 148 |
+
"loss": 1951.3572,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.08208255159474671,
|
| 153 |
+
"grad_norm": 9582.7724609375,
|
| 154 |
+
"learning_rate": 4.5937280050050835e-05,
|
| 155 |
+
"loss": 1944.6692,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.0859912445278299,
|
| 160 |
+
"grad_norm": 7010.27587890625,
|
| 161 |
+
"learning_rate": 4.574176898412451e-05,
|
| 162 |
+
"loss": 1955.3381,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08989993746091307,
|
| 167 |
+
"grad_norm": 7003.783203125,
|
| 168 |
+
"learning_rate": 4.554625791819817e-05,
|
| 169 |
+
"loss": 1949.6033,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.09380863039399624,
|
| 174 |
+
"grad_norm": 5024.95654296875,
|
| 175 |
+
"learning_rate": 4.5350746852271845e-05,
|
| 176 |
+
"loss": 1945.3767,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.09771732332707943,
|
| 181 |
+
"grad_norm": 9492.1552734375,
|
| 182 |
+
"learning_rate": 4.515523578634551e-05,
|
| 183 |
+
"loss": 1946.17,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.1016260162601626,
|
| 188 |
+
"grad_norm": 7467.62353515625,
|
| 189 |
+
"learning_rate": 4.4959724720419175e-05,
|
| 190 |
+
"loss": 1942.5891,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.10553470919324578,
|
| 195 |
+
"grad_norm": 3829.823974609375,
|
| 196 |
+
"learning_rate": 4.476421365449285e-05,
|
| 197 |
+
"loss": 1946.0959,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.10944340212632896,
|
| 202 |
+
"grad_norm": 4209.74853515625,
|
| 203 |
+
"learning_rate": 4.456870258856651e-05,
|
| 204 |
+
"loss": 1944.2936,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.11335209505941213,
|
| 209 |
+
"grad_norm": 4135.42626953125,
|
| 210 |
+
"learning_rate": 4.4373191522640185e-05,
|
| 211 |
+
"loss": 1946.382,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.11726078799249531,
|
| 216 |
+
"grad_norm": 9485.572265625,
|
| 217 |
+
"learning_rate": 4.417768045671386e-05,
|
| 218 |
+
"loss": 1942.6788,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.12116948092557848,
|
| 223 |
+
"grad_norm": 3942.744873046875,
|
| 224 |
+
"learning_rate": 4.398216939078752e-05,
|
| 225 |
+
"loss": 1940.9012,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.12507817385866166,
|
| 230 |
+
"grad_norm": 5323.8994140625,
|
| 231 |
+
"learning_rate": 4.378665832486119e-05,
|
| 232 |
+
"loss": 1941.8606,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.12898686679174484,
|
| 237 |
+
"grad_norm": 7794.04833984375,
|
| 238 |
+
"learning_rate": 4.3591147258934854e-05,
|
| 239 |
+
"loss": 1943.7128,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.13289555972482803,
|
| 244 |
+
"grad_norm": 5457.22265625,
|
| 245 |
+
"learning_rate": 4.3395636193008526e-05,
|
| 246 |
+
"loss": 1944.6231,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.13680425265791119,
|
| 251 |
+
"grad_norm": 5298.0380859375,
|
| 252 |
+
"learning_rate": 4.32001251270822e-05,
|
| 253 |
+
"loss": 1943.3045,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.14071294559099437,
|
| 258 |
+
"grad_norm": 6277.03759765625,
|
| 259 |
+
"learning_rate": 4.300461406115586e-05,
|
| 260 |
+
"loss": 1942.5687,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.14462163852407756,
|
| 265 |
+
"grad_norm": 6422.04052734375,
|
| 266 |
+
"learning_rate": 4.2809102995229536e-05,
|
| 267 |
+
"loss": 1942.5031,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1485303314571607,
|
| 272 |
+
"grad_norm": 4672.009765625,
|
| 273 |
+
"learning_rate": 4.26135919293032e-05,
|
| 274 |
+
"loss": 1935.7681,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1524390243902439,
|
| 279 |
+
"grad_norm": 9028.416015625,
|
| 280 |
+
"learning_rate": 4.2418080863376866e-05,
|
| 281 |
+
"loss": 1942.4916,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.15634771732332708,
|
| 286 |
+
"grad_norm": 2700.899169921875,
|
| 287 |
+
"learning_rate": 4.222256979745054e-05,
|
| 288 |
+
"loss": 1939.0205,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.16025641025641027,
|
| 293 |
+
"grad_norm": 5526.916015625,
|
| 294 |
+
"learning_rate": 4.202901384218347e-05,
|
| 295 |
+
"loss": 1941.2987,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.16416510318949343,
|
| 300 |
+
"grad_norm": 4574.68115234375,
|
| 301 |
+
"learning_rate": 4.183350277625714e-05,
|
| 302 |
+
"loss": 1937.8937,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1680737961225766,
|
| 307 |
+
"grad_norm": 1926.325439453125,
|
| 308 |
+
"learning_rate": 4.163799171033081e-05,
|
| 309 |
+
"loss": 1936.13,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.1719824890556598,
|
| 314 |
+
"grad_norm": 6505.677734375,
|
| 315 |
+
"learning_rate": 4.144248064440447e-05,
|
| 316 |
+
"loss": 1940.537,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.17589118198874296,
|
| 321 |
+
"grad_norm": 10048.75,
|
| 322 |
+
"learning_rate": 4.1246969578478145e-05,
|
| 323 |
+
"loss": 1937.288,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.17979987492182614,
|
| 328 |
+
"grad_norm": 5231.8671875,
|
| 329 |
+
"learning_rate": 4.105145851255181e-05,
|
| 330 |
+
"loss": 1937.2231,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.18370856785490933,
|
| 335 |
+
"grad_norm": 9982.53515625,
|
| 336 |
+
"learning_rate": 4.085594744662548e-05,
|
| 337 |
+
"loss": 1940.2748,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.18761726078799248,
|
| 342 |
+
"grad_norm": 8485.013671875,
|
| 343 |
+
"learning_rate": 4.0660436380699154e-05,
|
| 344 |
+
"loss": 1943.1405,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.19152595372107567,
|
| 349 |
+
"grad_norm": 11441.814453125,
|
| 350 |
+
"learning_rate": 4.046492531477282e-05,
|
| 351 |
+
"loss": 1943.1167,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.19543464665415886,
|
| 356 |
+
"grad_norm": 10675.1787109375,
|
| 357 |
+
"learning_rate": 4.0269414248846485e-05,
|
| 358 |
+
"loss": 1940.4123,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.199343339587242,
|
| 363 |
+
"grad_norm": 2382.636962890625,
|
| 364 |
+
"learning_rate": 4.007390318292015e-05,
|
| 365 |
+
"loss": 1943.3502,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.2032520325203252,
|
| 370 |
+
"grad_norm": 4759.23681640625,
|
| 371 |
+
"learning_rate": 3.987839211699382e-05,
|
| 372 |
+
"loss": 1937.6414,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.20716072545340838,
|
| 377 |
+
"grad_norm": 1630.03759765625,
|
| 378 |
+
"learning_rate": 3.9682881051067495e-05,
|
| 379 |
+
"loss": 1940.1084,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.21106941838649157,
|
| 384 |
+
"grad_norm": 6308.59375,
|
| 385 |
+
"learning_rate": 3.948736998514116e-05,
|
| 386 |
+
"loss": 1938.572,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.21497811131957473,
|
| 391 |
+
"grad_norm": 6484.08935546875,
|
| 392 |
+
"learning_rate": 3.929185891921483e-05,
|
| 393 |
+
"loss": 1937.7761,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2188868042526579,
|
| 398 |
+
"grad_norm": 5694.4111328125,
|
| 399 |
+
"learning_rate": 3.90963478532885e-05,
|
| 400 |
+
"loss": 1934.4069,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.2227954971857411,
|
| 405 |
+
"grad_norm": 4962.38623046875,
|
| 406 |
+
"learning_rate": 3.890083678736216e-05,
|
| 407 |
+
"loss": 1939.0358,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.22670419011882426,
|
| 412 |
+
"grad_norm": 6531.33056640625,
|
| 413 |
+
"learning_rate": 3.8705325721435836e-05,
|
| 414 |
+
"loss": 1936.8266,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.23061288305190744,
|
| 419 |
+
"grad_norm": 1153.397705078125,
|
| 420 |
+
"learning_rate": 3.85098146555095e-05,
|
| 421 |
+
"loss": 1934.8263,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.23452157598499063,
|
| 426 |
+
"grad_norm": 8893.7236328125,
|
| 427 |
+
"learning_rate": 3.831430358958317e-05,
|
| 428 |
+
"loss": 1939.5045,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.23843026891807378,
|
| 433 |
+
"grad_norm": 2473.042236328125,
|
| 434 |
+
"learning_rate": 3.8118792523656845e-05,
|
| 435 |
+
"loss": 1935.7797,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.24233896185115697,
|
| 440 |
+
"grad_norm": 3183.37451171875,
|
| 441 |
+
"learning_rate": 3.792523656838977e-05,
|
| 442 |
+
"loss": 1931.3295,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.24624765478424016,
|
| 447 |
+
"grad_norm": 9022.3369140625,
|
| 448 |
+
"learning_rate": 3.772972550246344e-05,
|
| 449 |
+
"loss": 1933.9825,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.2501563477173233,
|
| 454 |
+
"grad_norm": 5505.71337890625,
|
| 455 |
+
"learning_rate": 3.753421443653711e-05,
|
| 456 |
+
"loss": 1936.6508,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.2540650406504065,
|
| 461 |
+
"grad_norm": 6306.0283203125,
|
| 462 |
+
"learning_rate": 3.733870337061078e-05,
|
| 463 |
+
"loss": 1936.9334,
|
| 464 |
+
"step": 6500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.2579737335834897,
|
| 468 |
+
"grad_norm": 8688.0869140625,
|
| 469 |
+
"learning_rate": 3.714319230468445e-05,
|
| 470 |
+
"loss": 1938.5386,
|
| 471 |
+
"step": 6600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.26188242651657284,
|
| 475 |
+
"grad_norm": 4821.421875,
|
| 476 |
+
"learning_rate": 3.694768123875812e-05,
|
| 477 |
+
"loss": 1937.6906,
|
| 478 |
+
"step": 6700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.26579111944965605,
|
| 482 |
+
"grad_norm": 5253.9189453125,
|
| 483 |
+
"learning_rate": 3.675217017283178e-05,
|
| 484 |
+
"loss": 1938.4362,
|
| 485 |
+
"step": 6800
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2696998123827392,
|
| 489 |
+
"grad_norm": 9892.1142578125,
|
| 490 |
+
"learning_rate": 3.6556659106905454e-05,
|
| 491 |
+
"loss": 1939.7828,
|
| 492 |
+
"step": 6900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.27360850531582237,
|
| 496 |
+
"grad_norm": 7260.49755859375,
|
| 497 |
+
"learning_rate": 3.636114804097912e-05,
|
| 498 |
+
"loss": 1936.1073,
|
| 499 |
+
"step": 7000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2775171982489056,
|
| 503 |
+
"grad_norm": 8410.8330078125,
|
| 504 |
+
"learning_rate": 3.616563697505279e-05,
|
| 505 |
+
"loss": 1935.643,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.28142589118198874,
|
| 510 |
+
"grad_norm": 9192.7939453125,
|
| 511 |
+
"learning_rate": 3.597012590912646e-05,
|
| 512 |
+
"loss": 1937.7341,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.2853345841150719,
|
| 517 |
+
"grad_norm": 9378.2734375,
|
| 518 |
+
"learning_rate": 3.577461484320013e-05,
|
| 519 |
+
"loss": 1932.8997,
|
| 520 |
+
"step": 7300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2892432770481551,
|
| 524 |
+
"grad_norm": 3011.90771484375,
|
| 525 |
+
"learning_rate": 3.5579103777273795e-05,
|
| 526 |
+
"loss": 1931.0811,
|
| 527 |
+
"step": 7400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.29315196998123827,
|
| 531 |
+
"grad_norm": 8059.7666015625,
|
| 532 |
+
"learning_rate": 3.538359271134746e-05,
|
| 533 |
+
"loss": 1937.8091,
|
| 534 |
+
"step": 7500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.2970606629143214,
|
| 538 |
+
"grad_norm": 4335.9677734375,
|
| 539 |
+
"learning_rate": 3.518808164542113e-05,
|
| 540 |
+
"loss": 1934.6813,
|
| 541 |
+
"step": 7600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.30096935584740464,
|
| 545 |
+
"grad_norm": 3007.888916015625,
|
| 546 |
+
"learning_rate": 3.49925705794948e-05,
|
| 547 |
+
"loss": 1938.0166,
|
| 548 |
+
"step": 7700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.3048780487804878,
|
| 552 |
+
"grad_norm": 4845.6474609375,
|
| 553 |
+
"learning_rate": 3.479705951356847e-05,
|
| 554 |
+
"loss": 1932.5495,
|
| 555 |
+
"step": 7800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.30878674171357096,
|
| 559 |
+
"grad_norm": 4432.91796875,
|
| 560 |
+
"learning_rate": 3.460154844764214e-05,
|
| 561 |
+
"loss": 1930.5036,
|
| 562 |
+
"step": 7900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.31269543464665417,
|
| 566 |
+
"grad_norm": 4812.18994140625,
|
| 567 |
+
"learning_rate": 3.440603738171581e-05,
|
| 568 |
+
"loss": 1934.8019,
|
| 569 |
+
"step": 8000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.3166041275797373,
|
| 573 |
+
"grad_norm": 6923.658203125,
|
| 574 |
+
"learning_rate": 3.421052631578947e-05,
|
| 575 |
+
"loss": 1933.8223,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.32051282051282054,
|
| 580 |
+
"grad_norm": 5418.6611328125,
|
| 581 |
+
"learning_rate": 3.401501524986314e-05,
|
| 582 |
+
"loss": 1933.1703,
|
| 583 |
+
"step": 8200
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3244215134459037,
|
| 587 |
+
"grad_norm": 7177.35498046875,
|
| 588 |
+
"learning_rate": 3.3821459294596076e-05,
|
| 589 |
+
"loss": 1939.8072,
|
| 590 |
+
"step": 8300
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.32833020637898686,
|
| 594 |
+
"grad_norm": 4692.5712890625,
|
| 595 |
+
"learning_rate": 3.362594822866975e-05,
|
| 596 |
+
"loss": 1931.2592,
|
| 597 |
+
"step": 8400
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.33223889931207007,
|
| 601 |
+
"grad_norm": 5120.71240234375,
|
| 602 |
+
"learning_rate": 3.3430437162743414e-05,
|
| 603 |
+
"loss": 1930.9581,
|
| 604 |
+
"step": 8500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.3361475922451532,
|
| 608 |
+
"grad_norm": 7568.408203125,
|
| 609 |
+
"learning_rate": 3.323492609681708e-05,
|
| 610 |
+
"loss": 1934.3886,
|
| 611 |
+
"step": 8600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.3400562851782364,
|
| 615 |
+
"grad_norm": 5889.92138671875,
|
| 616 |
+
"learning_rate": 3.303941503089075e-05,
|
| 617 |
+
"loss": 1934.0802,
|
| 618 |
+
"step": 8700
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.3439649781113196,
|
| 622 |
+
"grad_norm": 4741.31494140625,
|
| 623 |
+
"learning_rate": 3.284390396496442e-05,
|
| 624 |
+
"loss": 1934.3195,
|
| 625 |
+
"step": 8800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.34787367104440275,
|
| 629 |
+
"grad_norm": 7635.93017578125,
|
| 630 |
+
"learning_rate": 3.264839289903809e-05,
|
| 631 |
+
"loss": 1934.2544,
|
| 632 |
+
"step": 8900
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3517823639774859,
|
| 636 |
+
"grad_norm": 3836.890869140625,
|
| 637 |
+
"learning_rate": 3.2452881833111754e-05,
|
| 638 |
+
"loss": 1933.2398,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3556910569105691,
|
| 643 |
+
"grad_norm": 7898.76220703125,
|
| 644 |
+
"learning_rate": 3.2257370767185427e-05,
|
| 645 |
+
"loss": 1936.3502,
|
| 646 |
+
"step": 9100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.3595997498436523,
|
| 650 |
+
"grad_norm": 8666.2724609375,
|
| 651 |
+
"learning_rate": 3.206185970125909e-05,
|
| 652 |
+
"loss": 1933.9322,
|
| 653 |
+
"step": 9200
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.36350844277673544,
|
| 657 |
+
"grad_norm": 2994.890625,
|
| 658 |
+
"learning_rate": 3.186634863533276e-05,
|
| 659 |
+
"loss": 1927.5716,
|
| 660 |
+
"step": 9300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.36741713570981865,
|
| 664 |
+
"grad_norm": 8034.71337890625,
|
| 665 |
+
"learning_rate": 3.167083756940643e-05,
|
| 666 |
+
"loss": 1933.1073,
|
| 667 |
+
"step": 9400
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.3713258286429018,
|
| 671 |
+
"grad_norm": 6698.51953125,
|
| 672 |
+
"learning_rate": 3.1475326503480095e-05,
|
| 673 |
+
"loss": 1929.7136,
|
| 674 |
+
"step": 9500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.37523452157598497,
|
| 678 |
+
"grad_norm": 6546.0263671875,
|
| 679 |
+
"learning_rate": 3.127981543755377e-05,
|
| 680 |
+
"loss": 1936.2655,
|
| 681 |
+
"step": 9600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3791432145090682,
|
| 685 |
+
"grad_norm": 3147.62646484375,
|
| 686 |
+
"learning_rate": 3.108430437162744e-05,
|
| 687 |
+
"loss": 1934.3759,
|
| 688 |
+
"step": 9700
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.38305190744215134,
|
| 692 |
+
"grad_norm": 8704.0126953125,
|
| 693 |
+
"learning_rate": 3.0888793305701105e-05,
|
| 694 |
+
"loss": 1934.7744,
|
| 695 |
+
"step": 9800
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3869606003752345,
|
| 699 |
+
"grad_norm": 3778.2109375,
|
| 700 |
+
"learning_rate": 3.069328223977477e-05,
|
| 701 |
+
"loss": 1933.9356,
|
| 702 |
+
"step": 9900
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3908692933083177,
|
| 706 |
+
"grad_norm": 3088.361328125,
|
| 707 |
+
"learning_rate": 3.0497771173848442e-05,
|
| 708 |
+
"loss": 1931.7581,
|
| 709 |
+
"step": 10000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.39477798624140087,
|
| 713 |
+
"grad_norm": 4705.4033203125,
|
| 714 |
+
"learning_rate": 3.0302260107922108e-05,
|
| 715 |
+
"loss": 1934.1992,
|
| 716 |
+
"step": 10100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.398686679174484,
|
| 720 |
+
"grad_norm": 8636.2822265625,
|
| 721 |
+
"learning_rate": 3.010674904199578e-05,
|
| 722 |
+
"loss": 1934.413,
|
| 723 |
+
"step": 10200
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.40259537210756724,
|
| 727 |
+
"grad_norm": 5006.08447265625,
|
| 728 |
+
"learning_rate": 2.9913193086728707e-05,
|
| 729 |
+
"loss": 1931.742,
|
| 730 |
+
"step": 10300
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.4065040650406504,
|
| 734 |
+
"grad_norm": 7133.03955078125,
|
| 735 |
+
"learning_rate": 2.971768202080238e-05,
|
| 736 |
+
"loss": 1932.8452,
|
| 737 |
+
"step": 10400
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.4104127579737336,
|
| 741 |
+
"grad_norm": 9219.9921875,
|
| 742 |
+
"learning_rate": 2.952217095487605e-05,
|
| 743 |
+
"loss": 1931.7769,
|
| 744 |
+
"step": 10500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.41432145090681677,
|
| 748 |
+
"grad_norm": 3849.6103515625,
|
| 749 |
+
"learning_rate": 2.9326659888949714e-05,
|
| 750 |
+
"loss": 1937.658,
|
| 751 |
+
"step": 10600
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.4182301438398999,
|
| 755 |
+
"grad_norm": 7912.2333984375,
|
| 756 |
+
"learning_rate": 2.9131148823023386e-05,
|
| 757 |
+
"loss": 1935.0505,
|
| 758 |
+
"step": 10700
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.42213883677298314,
|
| 762 |
+
"grad_norm": 3944.04833984375,
|
| 763 |
+
"learning_rate": 2.893563775709705e-05,
|
| 764 |
+
"loss": 1930.4119,
|
| 765 |
+
"step": 10800
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.4260475297060663,
|
| 769 |
+
"grad_norm": 5772.48193359375,
|
| 770 |
+
"learning_rate": 2.874012669117072e-05,
|
| 771 |
+
"loss": 1934.3783,
|
| 772 |
+
"step": 10900
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.42995622263914945,
|
| 776 |
+
"grad_norm": 7050.07080078125,
|
| 777 |
+
"learning_rate": 2.8544615625244392e-05,
|
| 778 |
+
"loss": 1936.8552,
|
| 779 |
+
"step": 11000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.43386491557223267,
|
| 783 |
+
"grad_norm": 4541.4365234375,
|
| 784 |
+
"learning_rate": 2.8349104559318058e-05,
|
| 785 |
+
"loss": 1936.5128,
|
| 786 |
+
"step": 11100
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.4377736085053158,
|
| 790 |
+
"grad_norm": 4451.1005859375,
|
| 791 |
+
"learning_rate": 2.8153593493391727e-05,
|
| 792 |
+
"loss": 1931.2498,
|
| 793 |
+
"step": 11200
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.441682301438399,
|
| 797 |
+
"grad_norm": 6433.82177734375,
|
| 798 |
+
"learning_rate": 2.79580824274654e-05,
|
| 799 |
+
"loss": 1933.7414,
|
| 800 |
+
"step": 11300
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4455909943714822,
|
| 804 |
+
"grad_norm": 8882.1875,
|
| 805 |
+
"learning_rate": 2.7762571361539064e-05,
|
| 806 |
+
"loss": 1933.9195,
|
| 807 |
+
"step": 11400
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.44949968730456535,
|
| 811 |
+
"grad_norm": 7860.77099609375,
|
| 812 |
+
"learning_rate": 2.7567060295612733e-05,
|
| 813 |
+
"loss": 1934.6942,
|
| 814 |
+
"step": 11500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.4534083802376485,
|
| 818 |
+
"grad_norm": 7216.9462890625,
|
| 819 |
+
"learning_rate": 2.73715492296864e-05,
|
| 820 |
+
"loss": 1934.0253,
|
| 821 |
+
"step": 11600
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.4573170731707317,
|
| 825 |
+
"grad_norm": 7442.84033203125,
|
| 826 |
+
"learning_rate": 2.717603816376007e-05,
|
| 827 |
+
"loss": 1929.6816,
|
| 828 |
+
"step": 11700
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.4612257661038149,
|
| 832 |
+
"grad_norm": 5934.62109375,
|
| 833 |
+
"learning_rate": 2.698052709783374e-05,
|
| 834 |
+
"loss": 1929.8956,
|
| 835 |
+
"step": 11800
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.46513445903689804,
|
| 839 |
+
"grad_norm": 5385.78662109375,
|
| 840 |
+
"learning_rate": 2.6785016031907405e-05,
|
| 841 |
+
"loss": 1931.7627,
|
| 842 |
+
"step": 11900
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.46904315196998125,
|
| 846 |
+
"grad_norm": 4509.1748046875,
|
| 847 |
+
"learning_rate": 2.6589504965981077e-05,
|
| 848 |
+
"loss": 1936.3969,
|
| 849 |
+
"step": 12000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.4729518449030644,
|
| 853 |
+
"grad_norm": 4036.121826171875,
|
| 854 |
+
"learning_rate": 2.6393993900054742e-05,
|
| 855 |
+
"loss": 1930.8831,
|
| 856 |
+
"step": 12100
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.47686053783614757,
|
| 860 |
+
"grad_norm": 4547.4462890625,
|
| 861 |
+
"learning_rate": 2.619848283412841e-05,
|
| 862 |
+
"loss": 1932.6319,
|
| 863 |
+
"step": 12200
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.4807692307692308,
|
| 867 |
+
"grad_norm": 5662.5234375,
|
| 868 |
+
"learning_rate": 2.6002971768202083e-05,
|
| 869 |
+
"loss": 1930.9198,
|
| 870 |
+
"step": 12300
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.48467792370231394,
|
| 874 |
+
"grad_norm": 9108.7509765625,
|
| 875 |
+
"learning_rate": 2.580941581293501e-05,
|
| 876 |
+
"loss": 1931.7472,
|
| 877 |
+
"step": 12400
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.4885866166353971,
|
| 881 |
+
"grad_norm": 3580.26904296875,
|
| 882 |
+
"learning_rate": 2.5613904747008683e-05,
|
| 883 |
+
"loss": 1934.1778,
|
| 884 |
+
"step": 12500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.4924953095684803,
|
| 888 |
+
"grad_norm": 1826.543212890625,
|
| 889 |
+
"learning_rate": 2.5418393681082352e-05,
|
| 890 |
+
"loss": 1933.6081,
|
| 891 |
+
"step": 12600
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.49640400250156347,
|
| 895 |
+
"grad_norm": 5028.68408203125,
|
| 896 |
+
"learning_rate": 2.5222882615156017e-05,
|
| 897 |
+
"loss": 1931.8184,
|
| 898 |
+
"step": 12700
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.5003126954346466,
|
| 902 |
+
"grad_norm": 4588.86376953125,
|
| 903 |
+
"learning_rate": 2.502737154922969e-05,
|
| 904 |
+
"loss": 1932.5213,
|
| 905 |
+
"step": 12800
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.5042213883677298,
|
| 909 |
+
"grad_norm": 5277.638671875,
|
| 910 |
+
"learning_rate": 2.4831860483303358e-05,
|
| 911 |
+
"loss": 1932.7469,
|
| 912 |
+
"step": 12900
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.508130081300813,
|
| 916 |
+
"grad_norm": 3516.3701171875,
|
| 917 |
+
"learning_rate": 2.4636349417377024e-05,
|
| 918 |
+
"loss": 1929.9806,
|
| 919 |
+
"step": 13000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.5120387742338962,
|
| 923 |
+
"grad_norm": 1964.23974609375,
|
| 924 |
+
"learning_rate": 2.4440838351450692e-05,
|
| 925 |
+
"loss": 1932.1733,
|
| 926 |
+
"step": 13100
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.5159474671669794,
|
| 930 |
+
"grad_norm": 5508.216796875,
|
| 931 |
+
"learning_rate": 2.424532728552436e-05,
|
| 932 |
+
"loss": 1931.7581,
|
| 933 |
+
"step": 13200
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.5198561601000625,
|
| 937 |
+
"grad_norm": 5174.5830078125,
|
| 938 |
+
"learning_rate": 2.404981621959803e-05,
|
| 939 |
+
"loss": 1932.5694,
|
| 940 |
+
"step": 13300
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.5237648530331457,
|
| 944 |
+
"grad_norm": 3573.059326171875,
|
| 945 |
+
"learning_rate": 2.38543051536717e-05,
|
| 946 |
+
"loss": 1932.6616,
|
| 947 |
+
"step": 13400
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5276735459662288,
|
| 951 |
+
"grad_norm": 3097.640869140625,
|
| 952 |
+
"learning_rate": 2.3658794087745368e-05,
|
| 953 |
+
"loss": 1929.3984,
|
| 954 |
+
"step": 13500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.5315822388993121,
|
| 958 |
+
"grad_norm": 1940.4635009765625,
|
| 959 |
+
"learning_rate": 2.3463283021819036e-05,
|
| 960 |
+
"loss": 1935.1516,
|
| 961 |
+
"step": 13600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5354909318323953,
|
| 965 |
+
"grad_norm": 5113.375,
|
| 966 |
+
"learning_rate": 2.3267771955892705e-05,
|
| 967 |
+
"loss": 1928.4305,
|
| 968 |
+
"step": 13700
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5393996247654784,
|
| 972 |
+
"grad_norm": 4622.5771484375,
|
| 973 |
+
"learning_rate": 2.3072260889966374e-05,
|
| 974 |
+
"loss": 1930.7387,
|
| 975 |
+
"step": 13800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.5433083176985616,
|
| 979 |
+
"grad_norm": 3968.635498046875,
|
| 980 |
+
"learning_rate": 2.2876749824040043e-05,
|
| 981 |
+
"loss": 1927.8052,
|
| 982 |
+
"step": 13900
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5472170106316447,
|
| 986 |
+
"grad_norm": 5931.80517578125,
|
| 987 |
+
"learning_rate": 2.2681238758113708e-05,
|
| 988 |
+
"loss": 1931.4133,
|
| 989 |
+
"step": 14000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.551125703564728,
|
| 993 |
+
"grad_norm": 2574.138427734375,
|
| 994 |
+
"learning_rate": 2.2485727692187377e-05,
|
| 995 |
+
"loss": 1934.4295,
|
| 996 |
+
"step": 14100
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5550343964978112,
|
| 1000 |
+
"grad_norm": 4328.2265625,
|
| 1001 |
+
"learning_rate": 2.229021662626105e-05,
|
| 1002 |
+
"loss": 1931.8806,
|
| 1003 |
+
"step": 14200
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5589430894308943,
|
| 1007 |
+
"grad_norm": 5181.68359375,
|
| 1008 |
+
"learning_rate": 2.2094705560334718e-05,
|
| 1009 |
+
"loss": 1931.9281,
|
| 1010 |
+
"step": 14300
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5628517823639775,
|
| 1014 |
+
"grad_norm": 4922.71875,
|
| 1015 |
+
"learning_rate": 2.1899194494408383e-05,
|
| 1016 |
+
"loss": 1930.5544,
|
| 1017 |
+
"step": 14400
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5667604752970606,
|
| 1021 |
+
"grad_norm": 8296.521484375,
|
| 1022 |
+
"learning_rate": 2.1703683428482052e-05,
|
| 1023 |
+
"loss": 1930.3347,
|
| 1024 |
+
"step": 14500
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5706691682301438,
|
| 1028 |
+
"grad_norm": 2995.131103515625,
|
| 1029 |
+
"learning_rate": 2.1510127473214986e-05,
|
| 1030 |
+
"loss": 1935.6934,
|
| 1031 |
+
"step": 14600
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.5745778611632271,
|
| 1035 |
+
"grad_norm": 4322.044921875,
|
| 1036 |
+
"learning_rate": 2.1314616407288655e-05,
|
| 1037 |
+
"loss": 1934.0597,
|
| 1038 |
+
"step": 14700
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5784865540963102,
|
| 1042 |
+
"grad_norm": 2032.8863525390625,
|
| 1043 |
+
"learning_rate": 2.111910534136232e-05,
|
| 1044 |
+
"loss": 1929.1128,
|
| 1045 |
+
"step": 14800
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5823952470293934,
|
| 1049 |
+
"grad_norm": 3598.51416015625,
|
| 1050 |
+
"learning_rate": 2.092359427543599e-05,
|
| 1051 |
+
"loss": 1927.9994,
|
| 1052 |
+
"step": 14900
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5863039399624765,
|
| 1056 |
+
"grad_norm": 5214.47119140625,
|
| 1057 |
+
"learning_rate": 2.072808320950966e-05,
|
| 1058 |
+
"loss": 1938.3528,
|
| 1059 |
+
"step": 15000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5902126328955597,
|
| 1063 |
+
"grad_norm": 2441.246337890625,
|
| 1064 |
+
"learning_rate": 2.0532572143583327e-05,
|
| 1065 |
+
"loss": 1935.7395,
|
| 1066 |
+
"step": 15100
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5941213258286429,
|
| 1070 |
+
"grad_norm": 7862.3271484375,
|
| 1071 |
+
"learning_rate": 2.0337061077656996e-05,
|
| 1072 |
+
"loss": 1936.9286,
|
| 1073 |
+
"step": 15200
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.5980300187617261,
|
| 1077 |
+
"grad_norm": 4373.83740234375,
|
| 1078 |
+
"learning_rate": 2.0141550011730665e-05,
|
| 1079 |
+
"loss": 1931.1372,
|
| 1080 |
+
"step": 15300
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.6019387116948093,
|
| 1084 |
+
"grad_norm": 4733.13623046875,
|
| 1085 |
+
"learning_rate": 1.9946038945804333e-05,
|
| 1086 |
+
"loss": 1929.0563,
|
| 1087 |
+
"step": 15400
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.6058474046278924,
|
| 1091 |
+
"grad_norm": 5266.6220703125,
|
| 1092 |
+
"learning_rate": 1.9750527879878002e-05,
|
| 1093 |
+
"loss": 1928.9506,
|
| 1094 |
+
"step": 15500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.6097560975609756,
|
| 1098 |
+
"grad_norm": 7364.53759765625,
|
| 1099 |
+
"learning_rate": 1.955501681395167e-05,
|
| 1100 |
+
"loss": 1929.6644,
|
| 1101 |
+
"step": 15600
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.6136647904940588,
|
| 1105 |
+
"grad_norm": 1991.6553955078125,
|
| 1106 |
+
"learning_rate": 1.935950574802534e-05,
|
| 1107 |
+
"loss": 1930.2667,
|
| 1108 |
+
"step": 15700
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.6175734834271419,
|
| 1112 |
+
"grad_norm": 3393.039306640625,
|
| 1113 |
+
"learning_rate": 1.9163994682099005e-05,
|
| 1114 |
+
"loss": 1930.2998,
|
| 1115 |
+
"step": 15800
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.6214821763602252,
|
| 1119 |
+
"grad_norm": 8285.298828125,
|
| 1120 |
+
"learning_rate": 1.8968483616172677e-05,
|
| 1121 |
+
"loss": 1931.7272,
|
| 1122 |
+
"step": 15900
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.6253908692933083,
|
| 1126 |
+
"grad_norm": 5009.90185546875,
|
| 1127 |
+
"learning_rate": 1.8772972550246346e-05,
|
| 1128 |
+
"loss": 1930.1348,
|
| 1129 |
+
"step": 16000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.6292995622263915,
|
| 1133 |
+
"grad_norm": 3938.380126953125,
|
| 1134 |
+
"learning_rate": 1.8577461484320015e-05,
|
| 1135 |
+
"loss": 1932.4734,
|
| 1136 |
+
"step": 16100
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.6332082551594747,
|
| 1140 |
+
"grad_norm": 3447.97412109375,
|
| 1141 |
+
"learning_rate": 1.838195041839368e-05,
|
| 1142 |
+
"loss": 1931.2173,
|
| 1143 |
+
"step": 16200
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.6371169480925578,
|
| 1147 |
+
"grad_norm": 2327.56396484375,
|
| 1148 |
+
"learning_rate": 1.818643935246735e-05,
|
| 1149 |
+
"loss": 1928.4989,
|
| 1150 |
+
"step": 16300
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.6410256410256411,
|
| 1154 |
+
"grad_norm": 4010.790771484375,
|
| 1155 |
+
"learning_rate": 1.799092828654102e-05,
|
| 1156 |
+
"loss": 1930.7745,
|
| 1157 |
+
"step": 16400
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6449343339587242,
|
| 1161 |
+
"grad_norm": 4573.65966796875,
|
| 1162 |
+
"learning_rate": 1.7795417220614687e-05,
|
| 1163 |
+
"loss": 1929.9284,
|
| 1164 |
+
"step": 16500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.6488430268918074,
|
| 1168 |
+
"grad_norm": 8766.54296875,
|
| 1169 |
+
"learning_rate": 1.7599906154688356e-05,
|
| 1170 |
+
"loss": 1933.8216,
|
| 1171 |
+
"step": 16600
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.6527517198248906,
|
| 1175 |
+
"grad_norm": 3133.171142578125,
|
| 1176 |
+
"learning_rate": 1.7404395088762024e-05,
|
| 1177 |
+
"loss": 1930.3795,
|
| 1178 |
+
"step": 16700
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6566604127579737,
|
| 1182 |
+
"grad_norm": 5405.63037109375,
|
| 1183 |
+
"learning_rate": 1.721083913349496e-05,
|
| 1184 |
+
"loss": 1931.9134,
|
| 1185 |
+
"step": 16800
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6605691056910569,
|
| 1189 |
+
"grad_norm": 9909.7265625,
|
| 1190 |
+
"learning_rate": 1.7015328067568624e-05,
|
| 1191 |
+
"loss": 1930.2619,
|
| 1192 |
+
"step": 16900
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6644777986241401,
|
| 1196 |
+
"grad_norm": 3389.27197265625,
|
| 1197 |
+
"learning_rate": 1.6819817001642293e-05,
|
| 1198 |
+
"loss": 1930.3102,
|
| 1199 |
+
"step": 17000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.6683864915572233,
|
| 1203 |
+
"grad_norm": 6031.01806640625,
|
| 1204 |
+
"learning_rate": 1.662430593571596e-05,
|
| 1205 |
+
"loss": 1932.8886,
|
| 1206 |
+
"step": 17100
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.6722951844903065,
|
| 1210 |
+
"grad_norm": 6059.17919921875,
|
| 1211 |
+
"learning_rate": 1.642879486978963e-05,
|
| 1212 |
+
"loss": 1932.06,
|
| 1213 |
+
"step": 17200
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6762038774233896,
|
| 1217 |
+
"grad_norm": 6261.0595703125,
|
| 1218 |
+
"learning_rate": 1.62332838038633e-05,
|
| 1219 |
+
"loss": 1929.5397,
|
| 1220 |
+
"step": 17300
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6801125703564728,
|
| 1224 |
+
"grad_norm": 3372.994873046875,
|
| 1225 |
+
"learning_rate": 1.6037772737936968e-05,
|
| 1226 |
+
"loss": 1932.5519,
|
| 1227 |
+
"step": 17400
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.6840212632895559,
|
| 1231 |
+
"grad_norm": 2616.115234375,
|
| 1232 |
+
"learning_rate": 1.5842261672010637e-05,
|
| 1233 |
+
"loss": 1931.5678,
|
| 1234 |
+
"step": 17500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.6879299562226392,
|
| 1238 |
+
"grad_norm": 1621.889892578125,
|
| 1239 |
+
"learning_rate": 1.5646750606084302e-05,
|
| 1240 |
+
"loss": 1930.0881,
|
| 1241 |
+
"step": 17600
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6918386491557224,
|
| 1245 |
+
"grad_norm": 5428.60595703125,
|
| 1246 |
+
"learning_rate": 1.5451239540157974e-05,
|
| 1247 |
+
"loss": 1932.4986,
|
| 1248 |
+
"step": 17700
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6957473420888055,
|
| 1252 |
+
"grad_norm": 4497.154296875,
|
| 1253 |
+
"learning_rate": 1.5255728474231643e-05,
|
| 1254 |
+
"loss": 1932.2756,
|
| 1255 |
+
"step": 17800
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6996560350218887,
|
| 1259 |
+
"grad_norm": 2205.099365234375,
|
| 1260 |
+
"learning_rate": 1.506021740830531e-05,
|
| 1261 |
+
"loss": 1929.9892,
|
| 1262 |
+
"step": 17900
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.7035647279549718,
|
| 1266 |
+
"grad_norm": 3068.186767578125,
|
| 1267 |
+
"learning_rate": 1.4864706342378979e-05,
|
| 1268 |
+
"loss": 1930.1973,
|
| 1269 |
+
"step": 18000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.707473420888055,
|
| 1273 |
+
"grad_norm": 4308.47607421875,
|
| 1274 |
+
"learning_rate": 1.466919527645265e-05,
|
| 1275 |
+
"loss": 1925.682,
|
| 1276 |
+
"step": 18100
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.7113821138211383,
|
| 1280 |
+
"grad_norm": 1013.2030029296875,
|
| 1281 |
+
"learning_rate": 1.4473684210526317e-05,
|
| 1282 |
+
"loss": 1928.9886,
|
| 1283 |
+
"step": 18200
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.7152908067542214,
|
| 1287 |
+
"grad_norm": 4193.64111328125,
|
| 1288 |
+
"learning_rate": 1.4278173144599986e-05,
|
| 1289 |
+
"loss": 1931.9095,
|
| 1290 |
+
"step": 18300
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.7191994996873046,
|
| 1294 |
+
"grad_norm": 6386.65966796875,
|
| 1295 |
+
"learning_rate": 1.4082662078673653e-05,
|
| 1296 |
+
"loss": 1926.7845,
|
| 1297 |
+
"step": 18400
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.7231081926203877,
|
| 1301 |
+
"grad_norm": 6890.64794921875,
|
| 1302 |
+
"learning_rate": 1.3887151012747321e-05,
|
| 1303 |
+
"loss": 1930.7619,
|
| 1304 |
+
"step": 18500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.7270168855534709,
|
| 1308 |
+
"grad_norm": 2349.202392578125,
|
| 1309 |
+
"learning_rate": 1.3691639946820992e-05,
|
| 1310 |
+
"loss": 1933.7834,
|
| 1311 |
+
"step": 18600
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.7309255784865542,
|
| 1315 |
+
"grad_norm": 6866.693359375,
|
| 1316 |
+
"learning_rate": 1.3496128880894659e-05,
|
| 1317 |
+
"loss": 1929.3033,
|
| 1318 |
+
"step": 18700
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.7348342714196373,
|
| 1322 |
+
"grad_norm": 1558.5277099609375,
|
| 1323 |
+
"learning_rate": 1.3300617814968328e-05,
|
| 1324 |
+
"loss": 1929.7986,
|
| 1325 |
+
"step": 18800
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.7387429643527205,
|
| 1329 |
+
"grad_norm": 3048.65966796875,
|
| 1330 |
+
"learning_rate": 1.3105106749041995e-05,
|
| 1331 |
+
"loss": 1930.0216,
|
| 1332 |
+
"step": 18900
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.7426516572858036,
|
| 1336 |
+
"grad_norm": 3468.932373046875,
|
| 1337 |
+
"learning_rate": 1.2909595683115665e-05,
|
| 1338 |
+
"loss": 1924.9683,
|
| 1339 |
+
"step": 19000
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.7465603502188868,
|
| 1343 |
+
"grad_norm": 7214.60595703125,
|
| 1344 |
+
"learning_rate": 1.2714084617189334e-05,
|
| 1345 |
+
"loss": 1932.2103,
|
| 1346 |
+
"step": 19100
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.7504690431519699,
|
| 1350 |
+
"grad_norm": 2444.197265625,
|
| 1351 |
+
"learning_rate": 1.2520528661922265e-05,
|
| 1352 |
+
"loss": 1928.3866,
|
| 1353 |
+
"step": 19200
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.7543777360850532,
|
| 1357 |
+
"grad_norm": 5131.85546875,
|
| 1358 |
+
"learning_rate": 1.2325017595995934e-05,
|
| 1359 |
+
"loss": 1931.5844,
|
| 1360 |
+
"step": 19300
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.7582864290181364,
|
| 1364 |
+
"grad_norm": 3939.27587890625,
|
| 1365 |
+
"learning_rate": 1.2129506530069603e-05,
|
| 1366 |
+
"loss": 1932.2191,
|
| 1367 |
+
"step": 19400
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.7621951219512195,
|
| 1371 |
+
"grad_norm": 3849.02294921875,
|
| 1372 |
+
"learning_rate": 1.1933995464143271e-05,
|
| 1373 |
+
"loss": 1931.0331,
|
| 1374 |
+
"step": 19500
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.7661038148843027,
|
| 1378 |
+
"grad_norm": 4608.310546875,
|
| 1379 |
+
"learning_rate": 1.173848439821694e-05,
|
| 1380 |
+
"loss": 1927.5684,
|
| 1381 |
+
"step": 19600
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.7700125078173858,
|
| 1385 |
+
"grad_norm": 1417.9674072265625,
|
| 1386 |
+
"learning_rate": 1.1542973332290607e-05,
|
| 1387 |
+
"loss": 1932.9777,
|
| 1388 |
+
"step": 19700
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.773921200750469,
|
| 1392 |
+
"grad_norm": 2343.1787109375,
|
| 1393 |
+
"learning_rate": 1.1347462266364276e-05,
|
| 1394 |
+
"loss": 1931.735,
|
| 1395 |
+
"step": 19800
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7778298936835523,
|
| 1399 |
+
"grad_norm": 4321.103515625,
|
| 1400 |
+
"learning_rate": 1.1151951200437945e-05,
|
| 1401 |
+
"loss": 1930.1662,
|
| 1402 |
+
"step": 19900
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7817385866166354,
|
| 1406 |
+
"grad_norm": 956.7057495117188,
|
| 1407 |
+
"learning_rate": 1.0956440134511614e-05,
|
| 1408 |
+
"loss": 1928.61,
|
| 1409 |
+
"step": 20000
|
| 1410 |
+
}
|
| 1411 |
+
],
|
| 1412 |
+
"logging_steps": 100,
|
| 1413 |
+
"max_steps": 25584,
|
| 1414 |
+
"num_input_tokens_seen": 0,
|
| 1415 |
+
"num_train_epochs": 1,
|
| 1416 |
+
"save_steps": 20000,
|
| 1417 |
+
"stateful_callbacks": {
|
| 1418 |
+
"TrainerControl": {
|
| 1419 |
+
"args": {
|
| 1420 |
+
"should_epoch_stop": false,
|
| 1421 |
+
"should_evaluate": false,
|
| 1422 |
+
"should_log": false,
|
| 1423 |
+
"should_save": true,
|
| 1424 |
+
"should_training_stop": false
|
| 1425 |
+
},
|
| 1426 |
+
"attributes": {}
|
| 1427 |
+
}
|
| 1428 |
+
},
|
| 1429 |
+
"total_flos": 1.404856958976e+16,
|
| 1430 |
+
"train_batch_size": 48,
|
| 1431 |
+
"trial_name": null,
|
| 1432 |
+
"trial_params": null
|
| 1433 |
+
}
|
checkpoint-20000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a49aab84d85040b6b09a345b8e1d86d1bff6e01f5aaeb435f5d3e9ae7ab13fe0
|
| 3 |
+
size 5240
|