Upload folder using huggingface_hub
Browse files- .gitattributes +2 -0
- checkpoint-13750/colbert_linear.pt +3 -0
- checkpoint-13750/config.json +28 -0
- checkpoint-13750/global_step13750/mp_rank_00_model_states.pt +3 -0
- checkpoint-13750/latest +1 -0
- checkpoint-13750/model.safetensors +3 -0
- checkpoint-13750/rng_state.pth +3 -0
- checkpoint-13750/sentencepiece.bpe.model +3 -0
- checkpoint-13750/sparse_linear.pt +3 -0
- checkpoint-13750/special_tokens_map.json +51 -0
- checkpoint-13750/tokenizer.json +3 -0
- checkpoint-13750/tokenizer_config.json +55 -0
- checkpoint-13750/trainer_state.json +1958 -0
- checkpoint-13750/training_args.bin +3 -0
- colbert_linear.pt +3 -0
- config.json +28 -0
- model.safetensors +3 -0
- sentencepiece.bpe.model +3 -0
- sparse_linear.pt +3 -0
- special_tokens_map.json +51 -0
- tokenizer.json +3 -0
- tokenizer_config.json +55 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
checkpoint-13750/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
checkpoint-13750/colbert_linear.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f7be00e4665ef50024732dbbb24a7c678ea315230a36f53ae35db8d84370f5ea
|
| 3 |
+
size 2100674
|
checkpoint-13750/config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/root/.cache/huggingface/hub/models--BAAI--bge-m3/snapshots/5617a9f61b028005a4858fdac845db406aefb181",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XLMRobertaModel"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 1024,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 4096,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 8194,
|
| 17 |
+
"model_type": "xlm-roberta",
|
| 18 |
+
"num_attention_heads": 16,
|
| 19 |
+
"num_hidden_layers": 24,
|
| 20 |
+
"output_past": true,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"torch_dtype": "float16",
|
| 24 |
+
"transformers_version": "4.44.2",
|
| 25 |
+
"type_vocab_size": 1,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 250002
|
| 28 |
+
}
|
checkpoint-13750/global_step13750/mp_rank_00_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d764e0348b2b0d3acb5a8c42b86a13bbfbd19d7f3a22220d078096111dd32acb
|
| 3 |
+
size 7963362556
|
checkpoint-13750/latest
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
global_step13750
|
checkpoint-13750/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:775c5e2cef13b1e7ea7008b79aa048c5b2291b91816c654d009d7900f68a1b25
|
| 3 |
+
size 1135554344
|
checkpoint-13750/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bc901ee3e03e0c508a5e98f7d099130c108df2d70439ada4bb38e0de25a5645f
|
| 3 |
+
size 14180
|
checkpoint-13750/sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
checkpoint-13750/sparse_linear.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:080af46a1cc532cb579e8a06bb66d1743f86d632b6c8c50b1dd4319b348002e5
|
| 3 |
+
size 3516
|
checkpoint-13750/special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
checkpoint-13750/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:33cd99e33ce09bdd8a6136fddfe90a1c47f85bafedf7309d0eecc19012d43586
|
| 3 |
+
size 17082897
|
checkpoint-13750/tokenizer_config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 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 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "<mask>",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"bos_token": "<s>",
|
| 45 |
+
"clean_up_tokenization_spaces": true,
|
| 46 |
+
"cls_token": "<s>",
|
| 47 |
+
"eos_token": "</s>",
|
| 48 |
+
"mask_token": "<mask>",
|
| 49 |
+
"model_max_length": 8192,
|
| 50 |
+
"pad_token": "<pad>",
|
| 51 |
+
"sep_token": "</s>",
|
| 52 |
+
"sp_model_kwargs": {},
|
| 53 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 54 |
+
"unk_token": "<unk>"
|
| 55 |
+
}
|
checkpoint-13750/trainer_state.json
ADDED
|
@@ -0,0 +1,1958 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 13750,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0036363636363636364,
|
| 13 |
+
"grad_norm": 8.730692897709844,
|
| 14 |
+
"learning_rate": 1.0827317615529767e-06,
|
| 15 |
+
"loss": 0.1943,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.007272727272727273,
|
| 20 |
+
"grad_norm": 6.563599993841062,
|
| 21 |
+
"learning_rate": 1.2745743112470354e-06,
|
| 22 |
+
"loss": 0.1472,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01090909090909091,
|
| 27 |
+
"grad_norm": 11.109271986600454,
|
| 28 |
+
"learning_rate": 1.386795008860794e-06,
|
| 29 |
+
"loss": 0.1713,
|
| 30 |
+
"step": 150
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.014545454545454545,
|
| 34 |
+
"grad_norm": 5.772674964443883,
|
| 35 |
+
"learning_rate": 1.4664168609410934e-06,
|
| 36 |
+
"loss": 0.1585,
|
| 37 |
+
"step": 200
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01818181818181818,
|
| 41 |
+
"grad_norm": 14.76596653457522,
|
| 42 |
+
"learning_rate": 1.5281763674824356e-06,
|
| 43 |
+
"loss": 0.1429,
|
| 44 |
+
"step": 250
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02181818181818182,
|
| 48 |
+
"grad_norm": 9.568061757019173,
|
| 49 |
+
"learning_rate": 1.5786375585548526e-06,
|
| 50 |
+
"loss": 0.1316,
|
| 51 |
+
"step": 300
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.025454545454545455,
|
| 55 |
+
"grad_norm": 10.301645195685637,
|
| 56 |
+
"learning_rate": 1.621301887696672e-06,
|
| 57 |
+
"loss": 0.1305,
|
| 58 |
+
"step": 350
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.02909090909090909,
|
| 62 |
+
"grad_norm": 5.6993526158027645,
|
| 63 |
+
"learning_rate": 1.6582594106351517e-06,
|
| 64 |
+
"loss": 0.109,
|
| 65 |
+
"step": 400
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03272727272727273,
|
| 69 |
+
"grad_norm": 4.915542646471226,
|
| 70 |
+
"learning_rate": 1.6908582561686116e-06,
|
| 71 |
+
"loss": 0.1269,
|
| 72 |
+
"step": 450
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03636363636363636,
|
| 76 |
+
"grad_norm": 9.392642904031657,
|
| 77 |
+
"learning_rate": 1.720018917176494e-06,
|
| 78 |
+
"loss": 0.1201,
|
| 79 |
+
"step": 500
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04,
|
| 83 |
+
"grad_norm": 5.047170718950901,
|
| 84 |
+
"learning_rate": 1.746397943764599e-06,
|
| 85 |
+
"loss": 0.1304,
|
| 86 |
+
"step": 550
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04363636363636364,
|
| 90 |
+
"grad_norm": 6.865869458534065,
|
| 91 |
+
"learning_rate": 1.7704801082489109e-06,
|
| 92 |
+
"loss": 0.1185,
|
| 93 |
+
"step": 600
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04727272727272727,
|
| 97 |
+
"grad_norm": 7.641476199550763,
|
| 98 |
+
"learning_rate": 1.7926335520703266e-06,
|
| 99 |
+
"loss": 0.1071,
|
| 100 |
+
"step": 650
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.05090909090909091,
|
| 104 |
+
"grad_norm": 5.401679074654614,
|
| 105 |
+
"learning_rate": 1.8131444373907304e-06,
|
| 106 |
+
"loss": 0.1153,
|
| 107 |
+
"step": 700
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05454545454545454,
|
| 111 |
+
"grad_norm": 6.576489460422398,
|
| 112 |
+
"learning_rate": 1.8322396147902533e-06,
|
| 113 |
+
"loss": 0.1104,
|
| 114 |
+
"step": 750
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05818181818181818,
|
| 118 |
+
"grad_norm": 7.069491006747404,
|
| 119 |
+
"learning_rate": 1.8501019603292104e-06,
|
| 120 |
+
"loss": 0.1164,
|
| 121 |
+
"step": 800
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06181818181818182,
|
| 125 |
+
"grad_norm": 5.9290634450105975,
|
| 126 |
+
"learning_rate": 1.866881054798162e-06,
|
| 127 |
+
"loss": 0.1219,
|
| 128 |
+
"step": 850
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06545454545454546,
|
| 132 |
+
"grad_norm": 5.426281987614212,
|
| 133 |
+
"learning_rate": 1.88270080586267e-06,
|
| 134 |
+
"loss": 0.1215,
|
| 135 |
+
"step": 900
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.06909090909090909,
|
| 139 |
+
"grad_norm": 7.8695827198040105,
|
| 140 |
+
"learning_rate": 1.8976650066475357e-06,
|
| 141 |
+
"loss": 0.1016,
|
| 142 |
+
"step": 950
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07272727272727272,
|
| 146 |
+
"grad_norm": 16.511242359906724,
|
| 147 |
+
"learning_rate": 1.9118614668705526e-06,
|
| 148 |
+
"loss": 0.1005,
|
| 149 |
+
"step": 1000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.07636363636363637,
|
| 153 |
+
"grad_norm": 5.456737119903023,
|
| 154 |
+
"learning_rate": 1.9253651350044893e-06,
|
| 155 |
+
"loss": 0.0967,
|
| 156 |
+
"step": 1050
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.08,
|
| 160 |
+
"grad_norm": 3.0750937470823105,
|
| 161 |
+
"learning_rate": 1.9382404934586575e-06,
|
| 162 |
+
"loss": 0.0989,
|
| 163 |
+
"step": 1100
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08363636363636363,
|
| 167 |
+
"grad_norm": 8.699529648266866,
|
| 168 |
+
"learning_rate": 1.9505434209019962e-06,
|
| 169 |
+
"loss": 0.0988,
|
| 170 |
+
"step": 1150
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08727272727272728,
|
| 174 |
+
"grad_norm": 6.77391900591751,
|
| 175 |
+
"learning_rate": 1.9623226579429693e-06,
|
| 176 |
+
"loss": 0.1078,
|
| 177 |
+
"step": 1200
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.09090909090909091,
|
| 181 |
+
"grad_norm": 6.290303703702585,
|
| 182 |
+
"learning_rate": 1.973620973411895e-06,
|
| 183 |
+
"loss": 0.1118,
|
| 184 |
+
"step": 1250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.09454545454545454,
|
| 188 |
+
"grad_norm": 4.7489952229331,
|
| 189 |
+
"learning_rate": 1.984476101764385e-06,
|
| 190 |
+
"loss": 0.1014,
|
| 191 |
+
"step": 1300
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.09818181818181818,
|
| 195 |
+
"grad_norm": 5.92022086118128,
|
| 196 |
+
"learning_rate": 1.9949215034764293e-06,
|
| 197 |
+
"loss": 0.0865,
|
| 198 |
+
"step": 1350
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.10181818181818182,
|
| 202 |
+
"grad_norm": 3.9298710865494013,
|
| 203 |
+
"learning_rate": 1.9961212121212122e-06,
|
| 204 |
+
"loss": 0.1019,
|
| 205 |
+
"step": 1400
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.10545454545454545,
|
| 209 |
+
"grad_norm": 1.35952985911884,
|
| 210 |
+
"learning_rate": 1.9880404040404038e-06,
|
| 211 |
+
"loss": 0.0889,
|
| 212 |
+
"step": 1450
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.10909090909090909,
|
| 216 |
+
"grad_norm": 6.074760388330947,
|
| 217 |
+
"learning_rate": 1.9799595959595958e-06,
|
| 218 |
+
"loss": 0.088,
|
| 219 |
+
"step": 1500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.11272727272727273,
|
| 223 |
+
"grad_norm": 3.281001635643186,
|
| 224 |
+
"learning_rate": 1.9718787878787877e-06,
|
| 225 |
+
"loss": 0.0939,
|
| 226 |
+
"step": 1550
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.11636363636363636,
|
| 230 |
+
"grad_norm": 2.834658163409011,
|
| 231 |
+
"learning_rate": 1.9637979797979797e-06,
|
| 232 |
+
"loss": 0.0941,
|
| 233 |
+
"step": 1600
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.12,
|
| 237 |
+
"grad_norm": 5.716408171893896,
|
| 238 |
+
"learning_rate": 1.9557171717171717e-06,
|
| 239 |
+
"loss": 0.0869,
|
| 240 |
+
"step": 1650
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.12363636363636364,
|
| 244 |
+
"grad_norm": 4.355380070215489,
|
| 245 |
+
"learning_rate": 1.9476363636363637e-06,
|
| 246 |
+
"loss": 0.0879,
|
| 247 |
+
"step": 1700
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.12727272727272726,
|
| 251 |
+
"grad_norm": 2.56506898084542,
|
| 252 |
+
"learning_rate": 1.939555555555555e-06,
|
| 253 |
+
"loss": 0.0833,
|
| 254 |
+
"step": 1750
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.13090909090909092,
|
| 258 |
+
"grad_norm": 3.1998963994898744,
|
| 259 |
+
"learning_rate": 1.9314747474747476e-06,
|
| 260 |
+
"loss": 0.0827,
|
| 261 |
+
"step": 1800
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.13454545454545455,
|
| 265 |
+
"grad_norm": 2.993946484706046,
|
| 266 |
+
"learning_rate": 1.923393939393939e-06,
|
| 267 |
+
"loss": 0.0884,
|
| 268 |
+
"step": 1850
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.13818181818181818,
|
| 272 |
+
"grad_norm": 4.066355592263713,
|
| 273 |
+
"learning_rate": 1.915313131313131e-06,
|
| 274 |
+
"loss": 0.0843,
|
| 275 |
+
"step": 1900
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.14181818181818182,
|
| 279 |
+
"grad_norm": 3.9444301109471422,
|
| 280 |
+
"learning_rate": 1.9072323232323231e-06,
|
| 281 |
+
"loss": 0.079,
|
| 282 |
+
"step": 1950
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.14545454545454545,
|
| 286 |
+
"grad_norm": 4.998815586950459,
|
| 287 |
+
"learning_rate": 1.899151515151515e-06,
|
| 288 |
+
"loss": 0.0742,
|
| 289 |
+
"step": 2000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.14909090909090908,
|
| 293 |
+
"grad_norm": 4.98600431497313,
|
| 294 |
+
"learning_rate": 1.891070707070707e-06,
|
| 295 |
+
"loss": 0.0883,
|
| 296 |
+
"step": 2050
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.15272727272727274,
|
| 300 |
+
"grad_norm": 7.068049048520739,
|
| 301 |
+
"learning_rate": 1.8829898989898988e-06,
|
| 302 |
+
"loss": 0.0888,
|
| 303 |
+
"step": 2100
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.15636363636363637,
|
| 307 |
+
"grad_norm": 6.445827098301751,
|
| 308 |
+
"learning_rate": 1.8749090909090908e-06,
|
| 309 |
+
"loss": 0.083,
|
| 310 |
+
"step": 2150
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.16,
|
| 314 |
+
"grad_norm": 4.422103821997445,
|
| 315 |
+
"learning_rate": 1.8668282828282826e-06,
|
| 316 |
+
"loss": 0.0784,
|
| 317 |
+
"step": 2200
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.16363636363636364,
|
| 321 |
+
"grad_norm": 10.593030877025758,
|
| 322 |
+
"learning_rate": 1.8589090909090909e-06,
|
| 323 |
+
"loss": 0.0907,
|
| 324 |
+
"step": 2250
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.16727272727272727,
|
| 328 |
+
"grad_norm": 5.73637966279623,
|
| 329 |
+
"learning_rate": 1.8508282828282828e-06,
|
| 330 |
+
"loss": 0.0745,
|
| 331 |
+
"step": 2300
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1709090909090909,
|
| 335 |
+
"grad_norm": 4.435065232821809,
|
| 336 |
+
"learning_rate": 1.8427474747474748e-06,
|
| 337 |
+
"loss": 0.0764,
|
| 338 |
+
"step": 2350
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17454545454545456,
|
| 342 |
+
"grad_norm": 6.932446538015379,
|
| 343 |
+
"learning_rate": 1.8346666666666666e-06,
|
| 344 |
+
"loss": 0.082,
|
| 345 |
+
"step": 2400
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1781818181818182,
|
| 349 |
+
"grad_norm": 8.365214538194435,
|
| 350 |
+
"learning_rate": 1.8265858585858586e-06,
|
| 351 |
+
"loss": 0.0792,
|
| 352 |
+
"step": 2450
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.18181818181818182,
|
| 356 |
+
"grad_norm": 3.8114318602084714,
|
| 357 |
+
"learning_rate": 1.8185050505050503e-06,
|
| 358 |
+
"loss": 0.0819,
|
| 359 |
+
"step": 2500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.18545454545454546,
|
| 363 |
+
"grad_norm": 8.151155598162092,
|
| 364 |
+
"learning_rate": 1.8104242424242425e-06,
|
| 365 |
+
"loss": 0.0796,
|
| 366 |
+
"step": 2550
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1890909090909091,
|
| 370 |
+
"grad_norm": 3.853054291632494,
|
| 371 |
+
"learning_rate": 1.8023434343434343e-06,
|
| 372 |
+
"loss": 0.0724,
|
| 373 |
+
"step": 2600
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.19272727272727272,
|
| 377 |
+
"grad_norm": 2.7845888134928454,
|
| 378 |
+
"learning_rate": 1.7942626262626262e-06,
|
| 379 |
+
"loss": 0.0778,
|
| 380 |
+
"step": 2650
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.19636363636363635,
|
| 384 |
+
"grad_norm": 6.178204576735234,
|
| 385 |
+
"learning_rate": 1.786181818181818e-06,
|
| 386 |
+
"loss": 0.0852,
|
| 387 |
+
"step": 2700
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.2,
|
| 391 |
+
"grad_norm": 3.2752914823154926,
|
| 392 |
+
"learning_rate": 1.7781010101010102e-06,
|
| 393 |
+
"loss": 0.0724,
|
| 394 |
+
"step": 2750
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.20363636363636364,
|
| 398 |
+
"grad_norm": 4.209509329574692,
|
| 399 |
+
"learning_rate": 1.770020202020202e-06,
|
| 400 |
+
"loss": 0.0793,
|
| 401 |
+
"step": 2800
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.20727272727272728,
|
| 405 |
+
"grad_norm": 6.139612323065608,
|
| 406 |
+
"learning_rate": 1.761939393939394e-06,
|
| 407 |
+
"loss": 0.07,
|
| 408 |
+
"step": 2850
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.2109090909090909,
|
| 412 |
+
"grad_norm": 2.455643448424175,
|
| 413 |
+
"learning_rate": 1.7538585858585857e-06,
|
| 414 |
+
"loss": 0.0738,
|
| 415 |
+
"step": 2900
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.21454545454545454,
|
| 419 |
+
"grad_norm": 5.4333976968022695,
|
| 420 |
+
"learning_rate": 1.7457777777777779e-06,
|
| 421 |
+
"loss": 0.0783,
|
| 422 |
+
"step": 2950
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.21818181818181817,
|
| 426 |
+
"grad_norm": 5.0280981201819,
|
| 427 |
+
"learning_rate": 1.7376969696969696e-06,
|
| 428 |
+
"loss": 0.0615,
|
| 429 |
+
"step": 3000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.22181818181818183,
|
| 433 |
+
"grad_norm": 6.437028886166945,
|
| 434 |
+
"learning_rate": 1.7296161616161616e-06,
|
| 435 |
+
"loss": 0.0766,
|
| 436 |
+
"step": 3050
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.22545454545454546,
|
| 440 |
+
"grad_norm": 2.5493783641885575,
|
| 441 |
+
"learning_rate": 1.7215353535353534e-06,
|
| 442 |
+
"loss": 0.0667,
|
| 443 |
+
"step": 3100
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.2290909090909091,
|
| 447 |
+
"grad_norm": 7.176307787653395,
|
| 448 |
+
"learning_rate": 1.7134545454545456e-06,
|
| 449 |
+
"loss": 0.0814,
|
| 450 |
+
"step": 3150
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.23272727272727273,
|
| 454 |
+
"grad_norm": 6.4838446342779585,
|
| 455 |
+
"learning_rate": 1.7053737373737373e-06,
|
| 456 |
+
"loss": 0.0804,
|
| 457 |
+
"step": 3200
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.23636363636363636,
|
| 461 |
+
"grad_norm": 6.05439389409086,
|
| 462 |
+
"learning_rate": 1.6972929292929293e-06,
|
| 463 |
+
"loss": 0.0758,
|
| 464 |
+
"step": 3250
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.24,
|
| 468 |
+
"grad_norm": 4.8914160530492925,
|
| 469 |
+
"learning_rate": 1.6893737373737374e-06,
|
| 470 |
+
"loss": 0.0726,
|
| 471 |
+
"step": 3300
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.24363636363636362,
|
| 475 |
+
"grad_norm": 5.1032181254180875,
|
| 476 |
+
"learning_rate": 1.6812929292929292e-06,
|
| 477 |
+
"loss": 0.0742,
|
| 478 |
+
"step": 3350
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.24727272727272728,
|
| 482 |
+
"grad_norm": 5.512044376565606,
|
| 483 |
+
"learning_rate": 1.6732121212121211e-06,
|
| 484 |
+
"loss": 0.0721,
|
| 485 |
+
"step": 3400
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2509090909090909,
|
| 489 |
+
"grad_norm": 6.215671409867116,
|
| 490 |
+
"learning_rate": 1.6651313131313131e-06,
|
| 491 |
+
"loss": 0.0714,
|
| 492 |
+
"step": 3450
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.2545454545454545,
|
| 496 |
+
"grad_norm": 4.934931799389331,
|
| 497 |
+
"learning_rate": 1.657050505050505e-06,
|
| 498 |
+
"loss": 0.0834,
|
| 499 |
+
"step": 3500
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2581818181818182,
|
| 503 |
+
"grad_norm": 5.766772809824982,
|
| 504 |
+
"learning_rate": 1.6489696969696969e-06,
|
| 505 |
+
"loss": 0.0719,
|
| 506 |
+
"step": 3550
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.26181818181818184,
|
| 510 |
+
"grad_norm": 2.3502736155525907,
|
| 511 |
+
"learning_rate": 1.6408888888888888e-06,
|
| 512 |
+
"loss": 0.0667,
|
| 513 |
+
"step": 3600
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.26545454545454544,
|
| 517 |
+
"grad_norm": 5.212095666658788,
|
| 518 |
+
"learning_rate": 1.6328080808080806e-06,
|
| 519 |
+
"loss": 0.0657,
|
| 520 |
+
"step": 3650
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2690909090909091,
|
| 524 |
+
"grad_norm": 2.2102868298776994,
|
| 525 |
+
"learning_rate": 1.6247272727272728e-06,
|
| 526 |
+
"loss": 0.0746,
|
| 527 |
+
"step": 3700
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.2727272727272727,
|
| 531 |
+
"grad_norm": 7.107367934409905,
|
| 532 |
+
"learning_rate": 1.6166464646464645e-06,
|
| 533 |
+
"loss": 0.0686,
|
| 534 |
+
"step": 3750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.27636363636363637,
|
| 538 |
+
"grad_norm": 4.9640350043256065,
|
| 539 |
+
"learning_rate": 1.6085656565656565e-06,
|
| 540 |
+
"loss": 0.0715,
|
| 541 |
+
"step": 3800
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.28,
|
| 545 |
+
"grad_norm": 3.8508170066035556,
|
| 546 |
+
"learning_rate": 1.6004848484848483e-06,
|
| 547 |
+
"loss": 0.0755,
|
| 548 |
+
"step": 3850
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.28363636363636363,
|
| 552 |
+
"grad_norm": 6.406204149035612,
|
| 553 |
+
"learning_rate": 1.5924040404040405e-06,
|
| 554 |
+
"loss": 0.0722,
|
| 555 |
+
"step": 3900
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.2872727272727273,
|
| 559 |
+
"grad_norm": 6.4755620057100005,
|
| 560 |
+
"learning_rate": 1.5843232323232322e-06,
|
| 561 |
+
"loss": 0.0653,
|
| 562 |
+
"step": 3950
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2909090909090909,
|
| 566 |
+
"grad_norm": 4.7041545767525985,
|
| 567 |
+
"learning_rate": 1.5762424242424242e-06,
|
| 568 |
+
"loss": 0.061,
|
| 569 |
+
"step": 4000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.29454545454545455,
|
| 573 |
+
"grad_norm": 6.597148868978393,
|
| 574 |
+
"learning_rate": 1.568161616161616e-06,
|
| 575 |
+
"loss": 0.0631,
|
| 576 |
+
"step": 4050
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.29818181818181816,
|
| 580 |
+
"grad_norm": 4.24188377459276,
|
| 581 |
+
"learning_rate": 1.5600808080808082e-06,
|
| 582 |
+
"loss": 0.0767,
|
| 583 |
+
"step": 4100
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3018181818181818,
|
| 587 |
+
"grad_norm": 5.185458229586008,
|
| 588 |
+
"learning_rate": 1.552e-06,
|
| 589 |
+
"loss": 0.0768,
|
| 590 |
+
"step": 4150
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3054545454545455,
|
| 594 |
+
"grad_norm": 7.282950927099741,
|
| 595 |
+
"learning_rate": 1.543919191919192e-06,
|
| 596 |
+
"loss": 0.0696,
|
| 597 |
+
"step": 4200
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.3090909090909091,
|
| 601 |
+
"grad_norm": 6.602554878720333,
|
| 602 |
+
"learning_rate": 1.5358383838383837e-06,
|
| 603 |
+
"loss": 0.0644,
|
| 604 |
+
"step": 4250
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.31272727272727274,
|
| 608 |
+
"grad_norm": 5.8760748346703755,
|
| 609 |
+
"learning_rate": 1.5277575757575756e-06,
|
| 610 |
+
"loss": 0.0629,
|
| 611 |
+
"step": 4300
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.31636363636363635,
|
| 615 |
+
"grad_norm": 5.87511103099038,
|
| 616 |
+
"learning_rate": 1.5198383838383837e-06,
|
| 617 |
+
"loss": 0.0699,
|
| 618 |
+
"step": 4350
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.32,
|
| 622 |
+
"grad_norm": 5.07793513016435,
|
| 623 |
+
"learning_rate": 1.5117575757575757e-06,
|
| 624 |
+
"loss": 0.0744,
|
| 625 |
+
"step": 4400
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.3236363636363636,
|
| 629 |
+
"grad_norm": 3.3381013307087595,
|
| 630 |
+
"learning_rate": 1.5036767676767677e-06,
|
| 631 |
+
"loss": 0.0671,
|
| 632 |
+
"step": 4450
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.32727272727272727,
|
| 636 |
+
"grad_norm": 6.087362688149139,
|
| 637 |
+
"learning_rate": 1.4955959595959594e-06,
|
| 638 |
+
"loss": 0.0687,
|
| 639 |
+
"step": 4500
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.33090909090909093,
|
| 643 |
+
"grad_norm": 3.459475929834813,
|
| 644 |
+
"learning_rate": 1.4875151515151514e-06,
|
| 645 |
+
"loss": 0.0685,
|
| 646 |
+
"step": 4550
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.33454545454545453,
|
| 650 |
+
"grad_norm": 4.891090834007153,
|
| 651 |
+
"learning_rate": 1.4794343434343432e-06,
|
| 652 |
+
"loss": 0.0725,
|
| 653 |
+
"step": 4600
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.3381818181818182,
|
| 657 |
+
"grad_norm": 4.691041549495357,
|
| 658 |
+
"learning_rate": 1.4713535353535354e-06,
|
| 659 |
+
"loss": 0.061,
|
| 660 |
+
"step": 4650
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3418181818181818,
|
| 664 |
+
"grad_norm": 5.924562513722335,
|
| 665 |
+
"learning_rate": 1.4632727272727271e-06,
|
| 666 |
+
"loss": 0.0758,
|
| 667 |
+
"step": 4700
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.34545454545454546,
|
| 671 |
+
"grad_norm": 4.580382327650832,
|
| 672 |
+
"learning_rate": 1.4551919191919191e-06,
|
| 673 |
+
"loss": 0.0743,
|
| 674 |
+
"step": 4750
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.3490909090909091,
|
| 678 |
+
"grad_norm": 4.6681099203941026,
|
| 679 |
+
"learning_rate": 1.4471111111111109e-06,
|
| 680 |
+
"loss": 0.0656,
|
| 681 |
+
"step": 4800
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3527272727272727,
|
| 685 |
+
"grad_norm": 5.030610606632313,
|
| 686 |
+
"learning_rate": 1.439030303030303e-06,
|
| 687 |
+
"loss": 0.0737,
|
| 688 |
+
"step": 4850
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.3563636363636364,
|
| 692 |
+
"grad_norm": 7.614614950202259,
|
| 693 |
+
"learning_rate": 1.4309494949494948e-06,
|
| 694 |
+
"loss": 0.0686,
|
| 695 |
+
"step": 4900
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.36,
|
| 699 |
+
"grad_norm": 5.981464365403329,
|
| 700 |
+
"learning_rate": 1.4228686868686868e-06,
|
| 701 |
+
"loss": 0.0748,
|
| 702 |
+
"step": 4950
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.36363636363636365,
|
| 706 |
+
"grad_norm": 4.277903564607442,
|
| 707 |
+
"learning_rate": 1.4147878787878786e-06,
|
| 708 |
+
"loss": 0.0568,
|
| 709 |
+
"step": 5000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.36727272727272725,
|
| 713 |
+
"grad_norm": 17.93564630692927,
|
| 714 |
+
"learning_rate": 1.4067070707070707e-06,
|
| 715 |
+
"loss": 0.0663,
|
| 716 |
+
"step": 5050
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.3709090909090909,
|
| 720 |
+
"grad_norm": 2.2042619733871374,
|
| 721 |
+
"learning_rate": 1.3986262626262625e-06,
|
| 722 |
+
"loss": 0.0648,
|
| 723 |
+
"step": 5100
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.37454545454545457,
|
| 727 |
+
"grad_norm": 4.712133353867323,
|
| 728 |
+
"learning_rate": 1.3905454545454545e-06,
|
| 729 |
+
"loss": 0.0773,
|
| 730 |
+
"step": 5150
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3781818181818182,
|
| 734 |
+
"grad_norm": 5.573958863346438,
|
| 735 |
+
"learning_rate": 1.3824646464646462e-06,
|
| 736 |
+
"loss": 0.0679,
|
| 737 |
+
"step": 5200
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.38181818181818183,
|
| 741 |
+
"grad_norm": 4.848377798083522,
|
| 742 |
+
"learning_rate": 1.3743838383838384e-06,
|
| 743 |
+
"loss": 0.0671,
|
| 744 |
+
"step": 5250
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.38545454545454544,
|
| 748 |
+
"grad_norm": 5.980348193388087,
|
| 749 |
+
"learning_rate": 1.3663030303030302e-06,
|
| 750 |
+
"loss": 0.0694,
|
| 751 |
+
"step": 5300
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3890909090909091,
|
| 755 |
+
"grad_norm": 2.672380009268972,
|
| 756 |
+
"learning_rate": 1.3582222222222222e-06,
|
| 757 |
+
"loss": 0.0676,
|
| 758 |
+
"step": 5350
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.3927272727272727,
|
| 762 |
+
"grad_norm": 2.964040453903747,
|
| 763 |
+
"learning_rate": 1.350141414141414e-06,
|
| 764 |
+
"loss": 0.0643,
|
| 765 |
+
"step": 5400
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.39636363636363636,
|
| 769 |
+
"grad_norm": 7.038083972721559,
|
| 770 |
+
"learning_rate": 1.342060606060606e-06,
|
| 771 |
+
"loss": 0.0632,
|
| 772 |
+
"step": 5450
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.4,
|
| 776 |
+
"grad_norm": 3.8224124558617496,
|
| 777 |
+
"learning_rate": 1.3339797979797979e-06,
|
| 778 |
+
"loss": 0.0731,
|
| 779 |
+
"step": 5500
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.4036363636363636,
|
| 783 |
+
"grad_norm": 7.048096138652444,
|
| 784 |
+
"learning_rate": 1.3258989898989899e-06,
|
| 785 |
+
"loss": 0.0594,
|
| 786 |
+
"step": 5550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.4072727272727273,
|
| 790 |
+
"grad_norm": 5.745560964316516,
|
| 791 |
+
"learning_rate": 1.3178181818181816e-06,
|
| 792 |
+
"loss": 0.0532,
|
| 793 |
+
"step": 5600
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.4109090909090909,
|
| 797 |
+
"grad_norm": 4.188291560272238,
|
| 798 |
+
"learning_rate": 1.3097373737373736e-06,
|
| 799 |
+
"loss": 0.0633,
|
| 800 |
+
"step": 5650
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.41454545454545455,
|
| 804 |
+
"grad_norm": 7.49562784549071,
|
| 805 |
+
"learning_rate": 1.3018181818181817e-06,
|
| 806 |
+
"loss": 0.0681,
|
| 807 |
+
"step": 5700
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.41818181818181815,
|
| 811 |
+
"grad_norm": 3.5275023161774404,
|
| 812 |
+
"learning_rate": 1.2937373737373737e-06,
|
| 813 |
+
"loss": 0.0664,
|
| 814 |
+
"step": 5750
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.4218181818181818,
|
| 818 |
+
"grad_norm": 4.119155442422377,
|
| 819 |
+
"learning_rate": 1.2856565656565656e-06,
|
| 820 |
+
"loss": 0.0617,
|
| 821 |
+
"step": 5800
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.4254545454545455,
|
| 825 |
+
"grad_norm": 3.065330909676055,
|
| 826 |
+
"learning_rate": 1.2775757575757576e-06,
|
| 827 |
+
"loss": 0.055,
|
| 828 |
+
"step": 5850
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.4290909090909091,
|
| 832 |
+
"grad_norm": 5.499516032340102,
|
| 833 |
+
"learning_rate": 1.2694949494949494e-06,
|
| 834 |
+
"loss": 0.0666,
|
| 835 |
+
"step": 5900
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.43272727272727274,
|
| 839 |
+
"grad_norm": 7.319764009851145,
|
| 840 |
+
"learning_rate": 1.2614141414141414e-06,
|
| 841 |
+
"loss": 0.0679,
|
| 842 |
+
"step": 5950
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.43636363636363634,
|
| 846 |
+
"grad_norm": 3.842328064446604,
|
| 847 |
+
"learning_rate": 1.2533333333333333e-06,
|
| 848 |
+
"loss": 0.0627,
|
| 849 |
+
"step": 6000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.44,
|
| 853 |
+
"grad_norm": 1.6429715042707909,
|
| 854 |
+
"learning_rate": 1.2452525252525253e-06,
|
| 855 |
+
"loss": 0.0566,
|
| 856 |
+
"step": 6050
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.44363636363636366,
|
| 860 |
+
"grad_norm": 3.3628676599920797,
|
| 861 |
+
"learning_rate": 1.237171717171717e-06,
|
| 862 |
+
"loss": 0.0678,
|
| 863 |
+
"step": 6100
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.44727272727272727,
|
| 867 |
+
"grad_norm": 2.2255401941396613,
|
| 868 |
+
"learning_rate": 1.229090909090909e-06,
|
| 869 |
+
"loss": 0.0536,
|
| 870 |
+
"step": 6150
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.4509090909090909,
|
| 874 |
+
"grad_norm": 4.667731753921009,
|
| 875 |
+
"learning_rate": 1.221010101010101e-06,
|
| 876 |
+
"loss": 0.0704,
|
| 877 |
+
"step": 6200
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.45454545454545453,
|
| 881 |
+
"grad_norm": 4.3551424870793465,
|
| 882 |
+
"learning_rate": 1.212929292929293e-06,
|
| 883 |
+
"loss": 0.0637,
|
| 884 |
+
"step": 6250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.4581818181818182,
|
| 888 |
+
"grad_norm": 3.3583233273895865,
|
| 889 |
+
"learning_rate": 1.2048484848484848e-06,
|
| 890 |
+
"loss": 0.0627,
|
| 891 |
+
"step": 6300
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.4618181818181818,
|
| 895 |
+
"grad_norm": 4.2019753851553165,
|
| 896 |
+
"learning_rate": 1.1967676767676767e-06,
|
| 897 |
+
"loss": 0.0549,
|
| 898 |
+
"step": 6350
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.46545454545454545,
|
| 902 |
+
"grad_norm": 2.5872795563931024,
|
| 903 |
+
"learning_rate": 1.1886868686868687e-06,
|
| 904 |
+
"loss": 0.0544,
|
| 905 |
+
"step": 6400
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4690909090909091,
|
| 909 |
+
"grad_norm": 7.041225020724765,
|
| 910 |
+
"learning_rate": 1.1806060606060607e-06,
|
| 911 |
+
"loss": 0.06,
|
| 912 |
+
"step": 6450
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.4727272727272727,
|
| 916 |
+
"grad_norm": 5.958896992660691,
|
| 917 |
+
"learning_rate": 1.1725252525252524e-06,
|
| 918 |
+
"loss": 0.0564,
|
| 919 |
+
"step": 6500
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.4763636363636364,
|
| 923 |
+
"grad_norm": 4.476780431745298,
|
| 924 |
+
"learning_rate": 1.1644444444444444e-06,
|
| 925 |
+
"loss": 0.053,
|
| 926 |
+
"step": 6550
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.48,
|
| 930 |
+
"grad_norm": 1.356776109954863,
|
| 931 |
+
"learning_rate": 1.1565252525252525e-06,
|
| 932 |
+
"loss": 0.0601,
|
| 933 |
+
"step": 6600
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.48363636363636364,
|
| 937 |
+
"grad_norm": 8.192732907739552,
|
| 938 |
+
"learning_rate": 1.1484444444444443e-06,
|
| 939 |
+
"loss": 0.0569,
|
| 940 |
+
"step": 6650
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.48727272727272725,
|
| 944 |
+
"grad_norm": 4.683769267327389,
|
| 945 |
+
"learning_rate": 1.1403636363636365e-06,
|
| 946 |
+
"loss": 0.0573,
|
| 947 |
+
"step": 6700
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.4909090909090909,
|
| 951 |
+
"grad_norm": 2.2927166931841096,
|
| 952 |
+
"learning_rate": 1.1322828282828282e-06,
|
| 953 |
+
"loss": 0.0555,
|
| 954 |
+
"step": 6750
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.49454545454545457,
|
| 958 |
+
"grad_norm": 5.921072631266765,
|
| 959 |
+
"learning_rate": 1.1242020202020202e-06,
|
| 960 |
+
"loss": 0.0751,
|
| 961 |
+
"step": 6800
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.49818181818181817,
|
| 965 |
+
"grad_norm": 6.639001409977233,
|
| 966 |
+
"learning_rate": 1.116121212121212e-06,
|
| 967 |
+
"loss": 0.0585,
|
| 968 |
+
"step": 6850
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5018181818181818,
|
| 972 |
+
"grad_norm": 4.5171318312031525,
|
| 973 |
+
"learning_rate": 1.108040404040404e-06,
|
| 974 |
+
"loss": 0.0598,
|
| 975 |
+
"step": 6900
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.5054545454545455,
|
| 979 |
+
"grad_norm": 3.889837713901535,
|
| 980 |
+
"learning_rate": 1.099959595959596e-06,
|
| 981 |
+
"loss": 0.0557,
|
| 982 |
+
"step": 6950
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.509090909090909,
|
| 986 |
+
"grad_norm": 4.7778306090713905,
|
| 987 |
+
"learning_rate": 1.091878787878788e-06,
|
| 988 |
+
"loss": 0.069,
|
| 989 |
+
"step": 7000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5127272727272727,
|
| 993 |
+
"grad_norm": 7.318064341364163,
|
| 994 |
+
"learning_rate": 1.0837979797979797e-06,
|
| 995 |
+
"loss": 0.0589,
|
| 996 |
+
"step": 7050
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5163636363636364,
|
| 1000 |
+
"grad_norm": 0.8134545807541753,
|
| 1001 |
+
"learning_rate": 1.0757171717171716e-06,
|
| 1002 |
+
"loss": 0.0568,
|
| 1003 |
+
"step": 7100
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.52,
|
| 1007 |
+
"grad_norm": 8.47016607740995,
|
| 1008 |
+
"learning_rate": 1.0676363636363636e-06,
|
| 1009 |
+
"loss": 0.065,
|
| 1010 |
+
"step": 7150
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5236363636363637,
|
| 1014 |
+
"grad_norm": 4.216730157024882,
|
| 1015 |
+
"learning_rate": 1.0595555555555556e-06,
|
| 1016 |
+
"loss": 0.0551,
|
| 1017 |
+
"step": 7200
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5272727272727272,
|
| 1021 |
+
"grad_norm": 8.409317988298476,
|
| 1022 |
+
"learning_rate": 1.0514747474747473e-06,
|
| 1023 |
+
"loss": 0.0591,
|
| 1024 |
+
"step": 7250
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5309090909090909,
|
| 1028 |
+
"grad_norm": 3.8221711861975938,
|
| 1029 |
+
"learning_rate": 1.0433939393939393e-06,
|
| 1030 |
+
"loss": 0.0652,
|
| 1031 |
+
"step": 7300
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.5345454545454545,
|
| 1035 |
+
"grad_norm": 5.941432574111638,
|
| 1036 |
+
"learning_rate": 1.0353131313131313e-06,
|
| 1037 |
+
"loss": 0.0555,
|
| 1038 |
+
"step": 7350
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5381818181818182,
|
| 1042 |
+
"grad_norm": 3.7955500798877027,
|
| 1043 |
+
"learning_rate": 1.0272323232323233e-06,
|
| 1044 |
+
"loss": 0.0639,
|
| 1045 |
+
"step": 7400
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5418181818181819,
|
| 1049 |
+
"grad_norm": 5.077036391729846,
|
| 1050 |
+
"learning_rate": 1.019151515151515e-06,
|
| 1051 |
+
"loss": 0.0574,
|
| 1052 |
+
"step": 7450
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5454545454545454,
|
| 1056 |
+
"grad_norm": 2.967208140674572,
|
| 1057 |
+
"learning_rate": 1.011070707070707e-06,
|
| 1058 |
+
"loss": 0.055,
|
| 1059 |
+
"step": 7500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5490909090909091,
|
| 1063 |
+
"grad_norm": 6.4064840134469065,
|
| 1064 |
+
"learning_rate": 1.002989898989899e-06,
|
| 1065 |
+
"loss": 0.0563,
|
| 1066 |
+
"step": 7550
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5527272727272727,
|
| 1070 |
+
"grad_norm": 8.224100322274762,
|
| 1071 |
+
"learning_rate": 9.949090909090908e-07,
|
| 1072 |
+
"loss": 0.0542,
|
| 1073 |
+
"step": 7600
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.5563636363636364,
|
| 1077 |
+
"grad_norm": 7.195740329809225,
|
| 1078 |
+
"learning_rate": 9.868282828282827e-07,
|
| 1079 |
+
"loss": 0.0518,
|
| 1080 |
+
"step": 7650
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.56,
|
| 1084 |
+
"grad_norm": 1.4648844395390421,
|
| 1085 |
+
"learning_rate": 9.787474747474747e-07,
|
| 1086 |
+
"loss": 0.0513,
|
| 1087 |
+
"step": 7700
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5636363636363636,
|
| 1091 |
+
"grad_norm": 1.2171654793506612,
|
| 1092 |
+
"learning_rate": 9.706666666666667e-07,
|
| 1093 |
+
"loss": 0.0611,
|
| 1094 |
+
"step": 7750
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5672727272727273,
|
| 1098 |
+
"grad_norm": 3.5164849479935687,
|
| 1099 |
+
"learning_rate": 9.625858585858584e-07,
|
| 1100 |
+
"loss": 0.0637,
|
| 1101 |
+
"step": 7800
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5709090909090909,
|
| 1105 |
+
"grad_norm": 7.553161151311784,
|
| 1106 |
+
"learning_rate": 9.545050505050504e-07,
|
| 1107 |
+
"loss": 0.0566,
|
| 1108 |
+
"step": 7850
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5745454545454546,
|
| 1112 |
+
"grad_norm": 5.745398629139221,
|
| 1113 |
+
"learning_rate": 9.464242424242424e-07,
|
| 1114 |
+
"loss": 0.065,
|
| 1115 |
+
"step": 7900
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5781818181818181,
|
| 1119 |
+
"grad_norm": 3.814619257014882,
|
| 1120 |
+
"learning_rate": 9.383434343434344e-07,
|
| 1121 |
+
"loss": 0.0598,
|
| 1122 |
+
"step": 7950
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5818181818181818,
|
| 1126 |
+
"grad_norm": 4.131828897750148,
|
| 1127 |
+
"learning_rate": 9.302626262626262e-07,
|
| 1128 |
+
"loss": 0.0712,
|
| 1129 |
+
"step": 8000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5854545454545454,
|
| 1133 |
+
"grad_norm": 3.4914242723281075,
|
| 1134 |
+
"learning_rate": 9.221818181818182e-07,
|
| 1135 |
+
"loss": 0.0608,
|
| 1136 |
+
"step": 8050
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5890909090909091,
|
| 1140 |
+
"grad_norm": 5.282146490628778,
|
| 1141 |
+
"learning_rate": 9.141010101010101e-07,
|
| 1142 |
+
"loss": 0.0555,
|
| 1143 |
+
"step": 8100
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5927272727272728,
|
| 1147 |
+
"grad_norm": 6.151329591511757,
|
| 1148 |
+
"learning_rate": 9.060202020202021e-07,
|
| 1149 |
+
"loss": 0.0597,
|
| 1150 |
+
"step": 8150
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5963636363636363,
|
| 1154 |
+
"grad_norm": 0.5491620334300515,
|
| 1155 |
+
"learning_rate": 8.979393939393939e-07,
|
| 1156 |
+
"loss": 0.0525,
|
| 1157 |
+
"step": 8200
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6,
|
| 1161 |
+
"grad_norm": 6.024173360125982,
|
| 1162 |
+
"learning_rate": 8.898585858585859e-07,
|
| 1163 |
+
"loss": 0.0622,
|
| 1164 |
+
"step": 8250
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.6036363636363636,
|
| 1168 |
+
"grad_norm": 7.030861805690174,
|
| 1169 |
+
"learning_rate": 8.817777777777778e-07,
|
| 1170 |
+
"loss": 0.0556,
|
| 1171 |
+
"step": 8300
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.6072727272727273,
|
| 1175 |
+
"grad_norm": 5.380176092336188,
|
| 1176 |
+
"learning_rate": 8.736969696969697e-07,
|
| 1177 |
+
"loss": 0.0542,
|
| 1178 |
+
"step": 8350
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.610909090909091,
|
| 1182 |
+
"grad_norm": 5.417266910258118,
|
| 1183 |
+
"learning_rate": 8.656161616161616e-07,
|
| 1184 |
+
"loss": 0.0492,
|
| 1185 |
+
"step": 8400
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6145454545454545,
|
| 1189 |
+
"grad_norm": 4.935151132976808,
|
| 1190 |
+
"learning_rate": 8.575353535353535e-07,
|
| 1191 |
+
"loss": 0.0492,
|
| 1192 |
+
"step": 8450
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6181818181818182,
|
| 1196 |
+
"grad_norm": 4.659946484892871,
|
| 1197 |
+
"learning_rate": 8.494545454545455e-07,
|
| 1198 |
+
"loss": 0.0571,
|
| 1199 |
+
"step": 8500
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.6218181818181818,
|
| 1203 |
+
"grad_norm": 6.771681262010034,
|
| 1204 |
+
"learning_rate": 8.413737373737373e-07,
|
| 1205 |
+
"loss": 0.0599,
|
| 1206 |
+
"step": 8550
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.6254545454545455,
|
| 1210 |
+
"grad_norm": 6.08702835668505,
|
| 1211 |
+
"learning_rate": 8.332929292929293e-07,
|
| 1212 |
+
"loss": 0.0685,
|
| 1213 |
+
"step": 8600
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6290909090909091,
|
| 1217 |
+
"grad_norm": 6.534364094809468,
|
| 1218 |
+
"learning_rate": 8.252121212121212e-07,
|
| 1219 |
+
"loss": 0.0546,
|
| 1220 |
+
"step": 8650
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6327272727272727,
|
| 1224 |
+
"grad_norm": 3.834673453780823,
|
| 1225 |
+
"learning_rate": 8.171313131313131e-07,
|
| 1226 |
+
"loss": 0.0597,
|
| 1227 |
+
"step": 8700
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.6363636363636364,
|
| 1231 |
+
"grad_norm": 6.698669517846524,
|
| 1232 |
+
"learning_rate": 8.09050505050505e-07,
|
| 1233 |
+
"loss": 0.056,
|
| 1234 |
+
"step": 8750
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.64,
|
| 1238 |
+
"grad_norm": 5.4043455793793775,
|
| 1239 |
+
"learning_rate": 8.00969696969697e-07,
|
| 1240 |
+
"loss": 0.0613,
|
| 1241 |
+
"step": 8800
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6436363636363637,
|
| 1245 |
+
"grad_norm": 1.660910042565409,
|
| 1246 |
+
"learning_rate": 7.928888888888889e-07,
|
| 1247 |
+
"loss": 0.0598,
|
| 1248 |
+
"step": 8850
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6472727272727272,
|
| 1252 |
+
"grad_norm": 7.795287679730981,
|
| 1253 |
+
"learning_rate": 7.848080808080808e-07,
|
| 1254 |
+
"loss": 0.0507,
|
| 1255 |
+
"step": 8900
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6509090909090909,
|
| 1259 |
+
"grad_norm": 3.293884222491616,
|
| 1260 |
+
"learning_rate": 7.767272727272727e-07,
|
| 1261 |
+
"loss": 0.0505,
|
| 1262 |
+
"step": 8950
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.6545454545454545,
|
| 1266 |
+
"grad_norm": 4.230324246940526,
|
| 1267 |
+
"learning_rate": 7.686464646464647e-07,
|
| 1268 |
+
"loss": 0.048,
|
| 1269 |
+
"step": 9000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.6581818181818182,
|
| 1273 |
+
"grad_norm": 3.140145943767063,
|
| 1274 |
+
"learning_rate": 7.605656565656565e-07,
|
| 1275 |
+
"loss": 0.0641,
|
| 1276 |
+
"step": 9050
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.6618181818181819,
|
| 1280 |
+
"grad_norm": 4.863206340412036,
|
| 1281 |
+
"learning_rate": 7.526464646464646e-07,
|
| 1282 |
+
"loss": 0.0598,
|
| 1283 |
+
"step": 9100
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.6654545454545454,
|
| 1287 |
+
"grad_norm": 3.458397202112644,
|
| 1288 |
+
"learning_rate": 7.445656565656565e-07,
|
| 1289 |
+
"loss": 0.0483,
|
| 1290 |
+
"step": 9150
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.6690909090909091,
|
| 1294 |
+
"grad_norm": 4.78959754830013,
|
| 1295 |
+
"learning_rate": 7.364848484848485e-07,
|
| 1296 |
+
"loss": 0.0536,
|
| 1297 |
+
"step": 9200
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.6727272727272727,
|
| 1301 |
+
"grad_norm": 10.740436647088291,
|
| 1302 |
+
"learning_rate": 7.284040404040404e-07,
|
| 1303 |
+
"loss": 0.0572,
|
| 1304 |
+
"step": 9250
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.6763636363636364,
|
| 1308 |
+
"grad_norm": 1.6489558557462445,
|
| 1309 |
+
"learning_rate": 7.204848484848484e-07,
|
| 1310 |
+
"loss": 0.0604,
|
| 1311 |
+
"step": 9300
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.68,
|
| 1315 |
+
"grad_norm": 7.3316901995966335,
|
| 1316 |
+
"learning_rate": 7.124040404040403e-07,
|
| 1317 |
+
"loss": 0.0559,
|
| 1318 |
+
"step": 9350
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6836363636363636,
|
| 1322 |
+
"grad_norm": 1.5208003767788665,
|
| 1323 |
+
"learning_rate": 7.043232323232323e-07,
|
| 1324 |
+
"loss": 0.0523,
|
| 1325 |
+
"step": 9400
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.6872727272727273,
|
| 1329 |
+
"grad_norm": 3.7519776533636673,
|
| 1330 |
+
"learning_rate": 6.962424242424242e-07,
|
| 1331 |
+
"loss": 0.0482,
|
| 1332 |
+
"step": 9450
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.6909090909090909,
|
| 1336 |
+
"grad_norm": 1.6222324278842861,
|
| 1337 |
+
"learning_rate": 6.881616161616161e-07,
|
| 1338 |
+
"loss": 0.0572,
|
| 1339 |
+
"step": 9500
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6945454545454546,
|
| 1343 |
+
"grad_norm": 7.0411497148292925,
|
| 1344 |
+
"learning_rate": 6.80080808080808e-07,
|
| 1345 |
+
"loss": 0.0571,
|
| 1346 |
+
"step": 9550
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.6981818181818182,
|
| 1350 |
+
"grad_norm": 2.6841788063694616,
|
| 1351 |
+
"learning_rate": 6.72e-07,
|
| 1352 |
+
"loss": 0.0564,
|
| 1353 |
+
"step": 9600
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.7018181818181818,
|
| 1357 |
+
"grad_norm": 6.819000782187993,
|
| 1358 |
+
"learning_rate": 6.639191919191919e-07,
|
| 1359 |
+
"loss": 0.0605,
|
| 1360 |
+
"step": 9650
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.7054545454545454,
|
| 1364 |
+
"grad_norm": 7.92210806178369,
|
| 1365 |
+
"learning_rate": 6.558383838383838e-07,
|
| 1366 |
+
"loss": 0.0548,
|
| 1367 |
+
"step": 9700
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.7090909090909091,
|
| 1371 |
+
"grad_norm": 4.811455749568285,
|
| 1372 |
+
"learning_rate": 6.477575757575757e-07,
|
| 1373 |
+
"loss": 0.0528,
|
| 1374 |
+
"step": 9750
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.7127272727272728,
|
| 1378 |
+
"grad_norm": 5.308321038455987,
|
| 1379 |
+
"learning_rate": 6.396767676767676e-07,
|
| 1380 |
+
"loss": 0.0527,
|
| 1381 |
+
"step": 9800
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.7163636363636363,
|
| 1385 |
+
"grad_norm": 5.069220520833826,
|
| 1386 |
+
"learning_rate": 6.315959595959595e-07,
|
| 1387 |
+
"loss": 0.0566,
|
| 1388 |
+
"step": 9850
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.72,
|
| 1392 |
+
"grad_norm": 5.151662369356777,
|
| 1393 |
+
"learning_rate": 6.235151515151514e-07,
|
| 1394 |
+
"loss": 0.0567,
|
| 1395 |
+
"step": 9900
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7236363636363636,
|
| 1399 |
+
"grad_norm": 6.063613002898996,
|
| 1400 |
+
"learning_rate": 6.154343434343434e-07,
|
| 1401 |
+
"loss": 0.0587,
|
| 1402 |
+
"step": 9950
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7272727272727273,
|
| 1406 |
+
"grad_norm": 1.0628386687029727,
|
| 1407 |
+
"learning_rate": 6.073535353535353e-07,
|
| 1408 |
+
"loss": 0.0539,
|
| 1409 |
+
"step": 10000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.730909090909091,
|
| 1413 |
+
"grad_norm": 4.641899576056589,
|
| 1414 |
+
"learning_rate": 5.992727272727272e-07,
|
| 1415 |
+
"loss": 0.0516,
|
| 1416 |
+
"step": 10050
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.7345454545454545,
|
| 1420 |
+
"grad_norm": 4.650551400255244,
|
| 1421 |
+
"learning_rate": 5.911919191919191e-07,
|
| 1422 |
+
"loss": 0.0471,
|
| 1423 |
+
"step": 10100
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.7381818181818182,
|
| 1427 |
+
"grad_norm": 4.567296054840706,
|
| 1428 |
+
"learning_rate": 5.831111111111111e-07,
|
| 1429 |
+
"loss": 0.0479,
|
| 1430 |
+
"step": 10150
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.7418181818181818,
|
| 1434 |
+
"grad_norm": 5.044836996902614,
|
| 1435 |
+
"learning_rate": 5.750303030303029e-07,
|
| 1436 |
+
"loss": 0.0561,
|
| 1437 |
+
"step": 10200
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.7454545454545455,
|
| 1441 |
+
"grad_norm": 4.248950379851986,
|
| 1442 |
+
"learning_rate": 5.669494949494949e-07,
|
| 1443 |
+
"loss": 0.0601,
|
| 1444 |
+
"step": 10250
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.7490909090909091,
|
| 1448 |
+
"grad_norm": 2.028141754034167,
|
| 1449 |
+
"learning_rate": 5.588686868686868e-07,
|
| 1450 |
+
"loss": 0.0422,
|
| 1451 |
+
"step": 10300
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.7527272727272727,
|
| 1455 |
+
"grad_norm": 10.007581888321717,
|
| 1456 |
+
"learning_rate": 5.507878787878788e-07,
|
| 1457 |
+
"loss": 0.0469,
|
| 1458 |
+
"step": 10350
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.7563636363636363,
|
| 1462 |
+
"grad_norm": 4.577989595773856,
|
| 1463 |
+
"learning_rate": 5.427070707070706e-07,
|
| 1464 |
+
"loss": 0.05,
|
| 1465 |
+
"step": 10400
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.76,
|
| 1469 |
+
"grad_norm": 6.521133984672062,
|
| 1470 |
+
"learning_rate": 5.346262626262626e-07,
|
| 1471 |
+
"loss": 0.0428,
|
| 1472 |
+
"step": 10450
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.7636363636363637,
|
| 1476 |
+
"grad_norm": 3.9509494509580705,
|
| 1477 |
+
"learning_rate": 5.265454545454545e-07,
|
| 1478 |
+
"loss": 0.0522,
|
| 1479 |
+
"step": 10500
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.7672727272727272,
|
| 1483 |
+
"grad_norm": 6.944726977217027,
|
| 1484 |
+
"learning_rate": 5.184646464646465e-07,
|
| 1485 |
+
"loss": 0.0515,
|
| 1486 |
+
"step": 10550
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.7709090909090909,
|
| 1490 |
+
"grad_norm": 3.831893622894902,
|
| 1491 |
+
"learning_rate": 5.103838383838383e-07,
|
| 1492 |
+
"loss": 0.0541,
|
| 1493 |
+
"step": 10600
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.7745454545454545,
|
| 1497 |
+
"grad_norm": 4.8538480771451695,
|
| 1498 |
+
"learning_rate": 5.023030303030302e-07,
|
| 1499 |
+
"loss": 0.06,
|
| 1500 |
+
"step": 10650
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.7781818181818182,
|
| 1504 |
+
"grad_norm": 9.741124343678399,
|
| 1505 |
+
"learning_rate": 4.942222222222222e-07,
|
| 1506 |
+
"loss": 0.0522,
|
| 1507 |
+
"step": 10700
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.7818181818181819,
|
| 1511 |
+
"grad_norm": 1.637375802142195,
|
| 1512 |
+
"learning_rate": 4.861414141414141e-07,
|
| 1513 |
+
"loss": 0.0479,
|
| 1514 |
+
"step": 10750
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.7854545454545454,
|
| 1518 |
+
"grad_norm": 2.3084795257142936,
|
| 1519 |
+
"learning_rate": 4.78060606060606e-07,
|
| 1520 |
+
"loss": 0.0626,
|
| 1521 |
+
"step": 10800
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.7890909090909091,
|
| 1525 |
+
"grad_norm": 6.674598298862028,
|
| 1526 |
+
"learning_rate": 4.69979797979798e-07,
|
| 1527 |
+
"loss": 0.0476,
|
| 1528 |
+
"step": 10850
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.7927272727272727,
|
| 1532 |
+
"grad_norm": 5.520833007044753,
|
| 1533 |
+
"learning_rate": 4.618989898989899e-07,
|
| 1534 |
+
"loss": 0.0473,
|
| 1535 |
+
"step": 10900
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.7963636363636364,
|
| 1539 |
+
"grad_norm": 7.535471892284138,
|
| 1540 |
+
"learning_rate": 4.5381818181818183e-07,
|
| 1541 |
+
"loss": 0.0541,
|
| 1542 |
+
"step": 10950
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.8,
|
| 1546 |
+
"grad_norm": 2.119845815305954,
|
| 1547 |
+
"learning_rate": 4.457373737373737e-07,
|
| 1548 |
+
"loss": 0.0469,
|
| 1549 |
+
"step": 11000
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.8036363636363636,
|
| 1553 |
+
"grad_norm": 4.865422744810339,
|
| 1554 |
+
"learning_rate": 4.376565656565656e-07,
|
| 1555 |
+
"loss": 0.0542,
|
| 1556 |
+
"step": 11050
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.8072727272727273,
|
| 1560 |
+
"grad_norm": 4.381485410017174,
|
| 1561 |
+
"learning_rate": 4.2957575757575754e-07,
|
| 1562 |
+
"loss": 0.0595,
|
| 1563 |
+
"step": 11100
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.8109090909090909,
|
| 1567 |
+
"grad_norm": 4.763162545803381,
|
| 1568 |
+
"learning_rate": 4.2149494949494947e-07,
|
| 1569 |
+
"loss": 0.0542,
|
| 1570 |
+
"step": 11150
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.8145454545454546,
|
| 1574 |
+
"grad_norm": 3.986784442089026,
|
| 1575 |
+
"learning_rate": 4.134141414141414e-07,
|
| 1576 |
+
"loss": 0.0533,
|
| 1577 |
+
"step": 11200
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.8181818181818182,
|
| 1581 |
+
"grad_norm": 1.9368964454962925,
|
| 1582 |
+
"learning_rate": 4.053333333333333e-07,
|
| 1583 |
+
"loss": 0.051,
|
| 1584 |
+
"step": 11250
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.8218181818181818,
|
| 1588 |
+
"grad_norm": 5.557279558773371,
|
| 1589 |
+
"learning_rate": 3.9725252525252523e-07,
|
| 1590 |
+
"loss": 0.0507,
|
| 1591 |
+
"step": 11300
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.8254545454545454,
|
| 1595 |
+
"grad_norm": 5.301180819559454,
|
| 1596 |
+
"learning_rate": 3.8917171717171715e-07,
|
| 1597 |
+
"loss": 0.0478,
|
| 1598 |
+
"step": 11350
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.8290909090909091,
|
| 1602 |
+
"grad_norm": 6.393245840471723,
|
| 1603 |
+
"learning_rate": 3.810909090909091e-07,
|
| 1604 |
+
"loss": 0.0541,
|
| 1605 |
+
"step": 11400
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.8327272727272728,
|
| 1609 |
+
"grad_norm": 5.140132428255607,
|
| 1610 |
+
"learning_rate": 3.731717171717171e-07,
|
| 1611 |
+
"loss": 0.0574,
|
| 1612 |
+
"step": 11450
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.8363636363636363,
|
| 1616 |
+
"grad_norm": 4.020573635641951,
|
| 1617 |
+
"learning_rate": 3.6509090909090904e-07,
|
| 1618 |
+
"loss": 0.0488,
|
| 1619 |
+
"step": 11500
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.84,
|
| 1623 |
+
"grad_norm": 8.669170555909107,
|
| 1624 |
+
"learning_rate": 3.5701010101010096e-07,
|
| 1625 |
+
"loss": 0.0459,
|
| 1626 |
+
"step": 11550
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.8436363636363636,
|
| 1630 |
+
"grad_norm": 2.7227046748798425,
|
| 1631 |
+
"learning_rate": 3.489292929292929e-07,
|
| 1632 |
+
"loss": 0.0427,
|
| 1633 |
+
"step": 11600
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.8472727272727273,
|
| 1637 |
+
"grad_norm": 8.264092633240582,
|
| 1638 |
+
"learning_rate": 3.408484848484848e-07,
|
| 1639 |
+
"loss": 0.0443,
|
| 1640 |
+
"step": 11650
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.850909090909091,
|
| 1644 |
+
"grad_norm": 2.2161705174100543,
|
| 1645 |
+
"learning_rate": 3.3276767676767673e-07,
|
| 1646 |
+
"loss": 0.05,
|
| 1647 |
+
"step": 11700
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.8545454545454545,
|
| 1651 |
+
"grad_norm": 4.049181895573167,
|
| 1652 |
+
"learning_rate": 3.2468686868686865e-07,
|
| 1653 |
+
"loss": 0.045,
|
| 1654 |
+
"step": 11750
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.8581818181818182,
|
| 1658 |
+
"grad_norm": 6.448219337405561,
|
| 1659 |
+
"learning_rate": 3.1660606060606057e-07,
|
| 1660 |
+
"loss": 0.0445,
|
| 1661 |
+
"step": 11800
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.8618181818181818,
|
| 1665 |
+
"grad_norm": 1.6198743531280058,
|
| 1666 |
+
"learning_rate": 3.0852525252525255e-07,
|
| 1667 |
+
"loss": 0.0436,
|
| 1668 |
+
"step": 11850
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.8654545454545455,
|
| 1672 |
+
"grad_norm": 5.887817361300306,
|
| 1673 |
+
"learning_rate": 3.0044444444444447e-07,
|
| 1674 |
+
"loss": 0.058,
|
| 1675 |
+
"step": 11900
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.8690909090909091,
|
| 1679 |
+
"grad_norm": 5.420521679930149,
|
| 1680 |
+
"learning_rate": 2.923636363636364e-07,
|
| 1681 |
+
"loss": 0.0578,
|
| 1682 |
+
"step": 11950
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.8727272727272727,
|
| 1686 |
+
"grad_norm": 6.9211836011036105,
|
| 1687 |
+
"learning_rate": 2.842828282828283e-07,
|
| 1688 |
+
"loss": 0.0468,
|
| 1689 |
+
"step": 12000
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.8763636363636363,
|
| 1693 |
+
"grad_norm": 6.962499989042162,
|
| 1694 |
+
"learning_rate": 2.7620202020202024e-07,
|
| 1695 |
+
"loss": 0.0565,
|
| 1696 |
+
"step": 12050
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.88,
|
| 1700 |
+
"grad_norm": 2.251073369325601,
|
| 1701 |
+
"learning_rate": 2.681212121212121e-07,
|
| 1702 |
+
"loss": 0.051,
|
| 1703 |
+
"step": 12100
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.8836363636363637,
|
| 1707 |
+
"grad_norm": 6.051100878218682,
|
| 1708 |
+
"learning_rate": 2.6004040404040403e-07,
|
| 1709 |
+
"loss": 0.0494,
|
| 1710 |
+
"step": 12150
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8872727272727273,
|
| 1714 |
+
"grad_norm": 5.559109629686928,
|
| 1715 |
+
"learning_rate": 2.5195959595959595e-07,
|
| 1716 |
+
"loss": 0.0508,
|
| 1717 |
+
"step": 12200
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.8909090909090909,
|
| 1721 |
+
"grad_norm": 2.7004728150520614,
|
| 1722 |
+
"learning_rate": 2.4387878787878787e-07,
|
| 1723 |
+
"loss": 0.0514,
|
| 1724 |
+
"step": 12250
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.8945454545454545,
|
| 1728 |
+
"grad_norm": 7.949939262110135,
|
| 1729 |
+
"learning_rate": 2.3579797979797977e-07,
|
| 1730 |
+
"loss": 0.0573,
|
| 1731 |
+
"step": 12300
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.8981818181818182,
|
| 1735 |
+
"grad_norm": 6.02491008010814,
|
| 1736 |
+
"learning_rate": 2.277171717171717e-07,
|
| 1737 |
+
"loss": 0.0526,
|
| 1738 |
+
"step": 12350
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.9018181818181819,
|
| 1742 |
+
"grad_norm": 2.893965651713638,
|
| 1743 |
+
"learning_rate": 2.196363636363636e-07,
|
| 1744 |
+
"loss": 0.0553,
|
| 1745 |
+
"step": 12400
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.9054545454545454,
|
| 1749 |
+
"grad_norm": 0.7646769278371894,
|
| 1750 |
+
"learning_rate": 2.1155555555555554e-07,
|
| 1751 |
+
"loss": 0.0485,
|
| 1752 |
+
"step": 12450
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.9090909090909091,
|
| 1756 |
+
"grad_norm": 2.4601955688086847,
|
| 1757 |
+
"learning_rate": 2.0347474747474746e-07,
|
| 1758 |
+
"loss": 0.0437,
|
| 1759 |
+
"step": 12500
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.9127272727272727,
|
| 1763 |
+
"grad_norm": 6.845607279107924,
|
| 1764 |
+
"learning_rate": 1.9539393939393938e-07,
|
| 1765 |
+
"loss": 0.0456,
|
| 1766 |
+
"step": 12550
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.9163636363636364,
|
| 1770 |
+
"grad_norm": 3.267607748623562,
|
| 1771 |
+
"learning_rate": 1.873131313131313e-07,
|
| 1772 |
+
"loss": 0.0451,
|
| 1773 |
+
"step": 12600
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.92,
|
| 1777 |
+
"grad_norm": 2.1934900287360666,
|
| 1778 |
+
"learning_rate": 1.7923232323232322e-07,
|
| 1779 |
+
"loss": 0.0461,
|
| 1780 |
+
"step": 12650
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.9236363636363636,
|
| 1784 |
+
"grad_norm": 2.090160880581652,
|
| 1785 |
+
"learning_rate": 1.7115151515151512e-07,
|
| 1786 |
+
"loss": 0.0505,
|
| 1787 |
+
"step": 12700
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.9272727272727272,
|
| 1791 |
+
"grad_norm": 4.966097333564902,
|
| 1792 |
+
"learning_rate": 1.6307070707070704e-07,
|
| 1793 |
+
"loss": 0.0537,
|
| 1794 |
+
"step": 12750
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.9309090909090909,
|
| 1798 |
+
"grad_norm": 2.1214979148493076,
|
| 1799 |
+
"learning_rate": 1.5498989898989896e-07,
|
| 1800 |
+
"loss": 0.0461,
|
| 1801 |
+
"step": 12800
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.9345454545454546,
|
| 1805 |
+
"grad_norm": 4.331745712735202,
|
| 1806 |
+
"learning_rate": 1.4690909090909089e-07,
|
| 1807 |
+
"loss": 0.0499,
|
| 1808 |
+
"step": 12850
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.9381818181818182,
|
| 1812 |
+
"grad_norm": 8.581243563077976,
|
| 1813 |
+
"learning_rate": 1.3882828282828284e-07,
|
| 1814 |
+
"loss": 0.0493,
|
| 1815 |
+
"step": 12900
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.9418181818181818,
|
| 1819 |
+
"grad_norm": 4.054442881590284,
|
| 1820 |
+
"learning_rate": 1.3074747474747476e-07,
|
| 1821 |
+
"loss": 0.0556,
|
| 1822 |
+
"step": 12950
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.9454545454545454,
|
| 1826 |
+
"grad_norm": 5.0357651442227285,
|
| 1827 |
+
"learning_rate": 1.2266666666666665e-07,
|
| 1828 |
+
"loss": 0.0523,
|
| 1829 |
+
"step": 13000
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.9490909090909091,
|
| 1833 |
+
"grad_norm": 4.173721876206746,
|
| 1834 |
+
"learning_rate": 1.1458585858585858e-07,
|
| 1835 |
+
"loss": 0.042,
|
| 1836 |
+
"step": 13050
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.9527272727272728,
|
| 1840 |
+
"grad_norm": 5.514831485888669,
|
| 1841 |
+
"learning_rate": 1.065050505050505e-07,
|
| 1842 |
+
"loss": 0.0492,
|
| 1843 |
+
"step": 13100
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.9563636363636364,
|
| 1847 |
+
"grad_norm": 5.89925297032685,
|
| 1848 |
+
"learning_rate": 9.842424242424242e-08,
|
| 1849 |
+
"loss": 0.0493,
|
| 1850 |
+
"step": 13150
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.96,
|
| 1854 |
+
"grad_norm": 6.457997696881727,
|
| 1855 |
+
"learning_rate": 9.034343434343433e-08,
|
| 1856 |
+
"loss": 0.0516,
|
| 1857 |
+
"step": 13200
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.9636363636363636,
|
| 1861 |
+
"grad_norm": 10.243171836638412,
|
| 1862 |
+
"learning_rate": 8.226262626262625e-08,
|
| 1863 |
+
"loss": 0.0529,
|
| 1864 |
+
"step": 13250
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.9672727272727273,
|
| 1868 |
+
"grad_norm": 2.4837463351233695,
|
| 1869 |
+
"learning_rate": 7.418181818181817e-08,
|
| 1870 |
+
"loss": 0.0418,
|
| 1871 |
+
"step": 13300
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.9709090909090909,
|
| 1875 |
+
"grad_norm": 7.115315330310798,
|
| 1876 |
+
"learning_rate": 6.61010101010101e-08,
|
| 1877 |
+
"loss": 0.0537,
|
| 1878 |
+
"step": 13350
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.9745454545454545,
|
| 1882 |
+
"grad_norm": 7.137360943243018,
|
| 1883 |
+
"learning_rate": 5.818181818181818e-08,
|
| 1884 |
+
"loss": 0.0446,
|
| 1885 |
+
"step": 13400
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.9781818181818182,
|
| 1889 |
+
"grad_norm": 5.279850881667171,
|
| 1890 |
+
"learning_rate": 5.0101010101010105e-08,
|
| 1891 |
+
"loss": 0.0504,
|
| 1892 |
+
"step": 13450
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.9818181818181818,
|
| 1896 |
+
"grad_norm": 4.781435096498879,
|
| 1897 |
+
"learning_rate": 4.202020202020202e-08,
|
| 1898 |
+
"loss": 0.0519,
|
| 1899 |
+
"step": 13500
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.9854545454545455,
|
| 1903 |
+
"grad_norm": 4.141078938486959,
|
| 1904 |
+
"learning_rate": 3.393939393939394e-08,
|
| 1905 |
+
"loss": 0.057,
|
| 1906 |
+
"step": 13550
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.9890909090909091,
|
| 1910 |
+
"grad_norm": 3.983226535807257,
|
| 1911 |
+
"learning_rate": 2.5858585858585858e-08,
|
| 1912 |
+
"loss": 0.0588,
|
| 1913 |
+
"step": 13600
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.9927272727272727,
|
| 1917 |
+
"grad_norm": 8.448308055260652,
|
| 1918 |
+
"learning_rate": 1.7777777777777777e-08,
|
| 1919 |
+
"loss": 0.0578,
|
| 1920 |
+
"step": 13650
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.9963636363636363,
|
| 1924 |
+
"grad_norm": 2.0151546432182283,
|
| 1925 |
+
"learning_rate": 9.696969696969696e-09,
|
| 1926 |
+
"loss": 0.0503,
|
| 1927 |
+
"step": 13700
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 1.0,
|
| 1931 |
+
"grad_norm": 2.6109640799264664,
|
| 1932 |
+
"learning_rate": 1.6161616161616161e-09,
|
| 1933 |
+
"loss": 0.0573,
|
| 1934 |
+
"step": 13750
|
| 1935 |
+
}
|
| 1936 |
+
],
|
| 1937 |
+
"logging_steps": 50,
|
| 1938 |
+
"max_steps": 13750,
|
| 1939 |
+
"num_input_tokens_seen": 0,
|
| 1940 |
+
"num_train_epochs": 1,
|
| 1941 |
+
"save_steps": 3000,
|
| 1942 |
+
"stateful_callbacks": {
|
| 1943 |
+
"TrainerControl": {
|
| 1944 |
+
"args": {
|
| 1945 |
+
"should_epoch_stop": false,
|
| 1946 |
+
"should_evaluate": false,
|
| 1947 |
+
"should_log": false,
|
| 1948 |
+
"should_save": true,
|
| 1949 |
+
"should_training_stop": true
|
| 1950 |
+
},
|
| 1951 |
+
"attributes": {}
|
| 1952 |
+
}
|
| 1953 |
+
},
|
| 1954 |
+
"total_flos": 0.0,
|
| 1955 |
+
"train_batch_size": 1,
|
| 1956 |
+
"trial_name": null,
|
| 1957 |
+
"trial_params": null
|
| 1958 |
+
}
|
checkpoint-13750/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:285bfdca5f2c1cbaed9b426a316da5bce893571d9889fdd0efb10d0e6f5c360f
|
| 3 |
+
size 7416
|
colbert_linear.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f7be00e4665ef50024732dbbb24a7c678ea315230a36f53ae35db8d84370f5ea
|
| 3 |
+
size 2100674
|
config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/root/.cache/huggingface/hub/models--BAAI--bge-m3/snapshots/5617a9f61b028005a4858fdac845db406aefb181",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XLMRobertaModel"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 1024,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 4096,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 8194,
|
| 17 |
+
"model_type": "xlm-roberta",
|
| 18 |
+
"num_attention_heads": 16,
|
| 19 |
+
"num_hidden_layers": 24,
|
| 20 |
+
"output_past": true,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"torch_dtype": "float16",
|
| 24 |
+
"transformers_version": "4.44.2",
|
| 25 |
+
"type_vocab_size": 1,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 250002
|
| 28 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:775c5e2cef13b1e7ea7008b79aa048c5b2291b91816c654d009d7900f68a1b25
|
| 3 |
+
size 1135554344
|
sentencepiece.bpe.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
| 3 |
+
size 5069051
|
sparse_linear.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:080af46a1cc532cb579e8a06bb66d1743f86d632b6c8c50b1dd4319b348002e5
|
| 3 |
+
size 3516
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:33cd99e33ce09bdd8a6136fddfe90a1c47f85bafedf7309d0eecc19012d43586
|
| 3 |
+
size 17082897
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<pad>",
|
| 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 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"250001": {
|
| 36 |
+
"content": "<mask>",
|
| 37 |
+
"lstrip": true,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"bos_token": "<s>",
|
| 45 |
+
"clean_up_tokenization_spaces": true,
|
| 46 |
+
"cls_token": "<s>",
|
| 47 |
+
"eos_token": "</s>",
|
| 48 |
+
"mask_token": "<mask>",
|
| 49 |
+
"model_max_length": 8192,
|
| 50 |
+
"pad_token": "<pad>",
|
| 51 |
+
"sep_token": "</s>",
|
| 52 |
+
"sp_model_kwargs": {},
|
| 53 |
+
"tokenizer_class": "XLMRobertaTokenizer",
|
| 54 |
+
"unk_token": "<unk>"
|
| 55 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:285bfdca5f2c1cbaed9b426a316da5bce893571d9889fdd0efb10d0e6f5c360f
|
| 3 |
+
size 7416
|