Upload folder using huggingface_hub
Browse files- checkpoint-2000/config.json +25 -0
- checkpoint-2000/generation_config.json +5 -0
- checkpoint-2000/model.safetensors +3 -0
- checkpoint-2000/optimizer.pt +3 -0
- checkpoint-2000/rng_state.pth +3 -0
- checkpoint-2000/scheduler.pt +3 -0
- checkpoint-2000/trainer_state.json +1449 -0
- checkpoint-2000/training_args.bin +3 -0
- checkpoint-2298/config.json +25 -0
- checkpoint-2298/generation_config.json +5 -0
- checkpoint-2298/model.safetensors +3 -0
- checkpoint-2298/optimizer.pt +3 -0
- checkpoint-2298/rng_state.pth +3 -0
- checkpoint-2298/scheduler.pt +3 -0
- checkpoint-2298/trainer_state.json +1652 -0
- checkpoint-2298/training_args.bin +3 -0
- config.json +25 -0
- generation_config.json +5 -0
- model.safetensors +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +59 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
checkpoint-2000/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "dbmdz/bert-base-turkish-uncased",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"torch_dtype": "float32",
|
| 21 |
+
"transformers_version": "4.48.2",
|
| 22 |
+
"type_vocab_size": 2,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 32000
|
| 25 |
+
}
|
checkpoint-2000/generation_config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"pad_token_id": 0,
|
| 4 |
+
"transformers_version": "4.48.2"
|
| 5 |
+
}
|
checkpoint-2000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6525b9cf94979c93dbc724ffe8422ed123256dc5e96c9ddaa62e3bdc5e1f9ce8
|
| 3 |
+
size 442627224
|
checkpoint-2000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d23fd79d86ac1eb9a7f2be99e7752db70880364346cc6593748dda84169edbfa
|
| 3 |
+
size 885375994
|
checkpoint-2000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:96f5522a96bee3838e2f15b15c0498da86c5b5c1f63b681dd7f6e1b867c81359
|
| 3 |
+
size 14244
|
checkpoint-2000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cae0652f1824f4b8f92c162a19f1be78622a640efea9aa1b23750526d89c7e68
|
| 3 |
+
size 1064
|
checkpoint-2000/trainer_state.json
ADDED
|
@@ -0,0 +1,1449 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.6109660574412534,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.013054830287206266,
|
| 13 |
+
"grad_norm": 43.415225982666016,
|
| 14 |
+
"learning_rate": 4.978241949521324e-05,
|
| 15 |
+
"loss": 2.9884,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02610966057441253,
|
| 20 |
+
"grad_norm": 49.437705993652344,
|
| 21 |
+
"learning_rate": 4.956483899042646e-05,
|
| 22 |
+
"loss": 2.9654,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0391644908616188,
|
| 27 |
+
"grad_norm": 47.24225997924805,
|
| 28 |
+
"learning_rate": 4.934725848563969e-05,
|
| 29 |
+
"loss": 3.5102,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.05221932114882506,
|
| 34 |
+
"grad_norm": 32.93499755859375,
|
| 35 |
+
"learning_rate": 4.912967798085292e-05,
|
| 36 |
+
"loss": 2.4143,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.06527415143603134,
|
| 41 |
+
"grad_norm": 31.01405143737793,
|
| 42 |
+
"learning_rate": 4.891209747606615e-05,
|
| 43 |
+
"loss": 2.6864,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0783289817232376,
|
| 48 |
+
"grad_norm": 77.32862854003906,
|
| 49 |
+
"learning_rate": 4.8694516971279375e-05,
|
| 50 |
+
"loss": 2.7403,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.09138381201044386,
|
| 55 |
+
"grad_norm": 28.313295364379883,
|
| 56 |
+
"learning_rate": 4.84769364664926e-05,
|
| 57 |
+
"loss": 2.5107,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.10443864229765012,
|
| 62 |
+
"grad_norm": 38.45579528808594,
|
| 63 |
+
"learning_rate": 4.825935596170583e-05,
|
| 64 |
+
"loss": 2.6546,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1174934725848564,
|
| 69 |
+
"grad_norm": 41.91643142700195,
|
| 70 |
+
"learning_rate": 4.8041775456919065e-05,
|
| 71 |
+
"loss": 2.82,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.13054830287206268,
|
| 76 |
+
"grad_norm": 36.32301712036133,
|
| 77 |
+
"learning_rate": 4.782419495213229e-05,
|
| 78 |
+
"loss": 2.9063,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.14360313315926893,
|
| 83 |
+
"grad_norm": 30.05735969543457,
|
| 84 |
+
"learning_rate": 4.760661444734552e-05,
|
| 85 |
+
"loss": 2.5942,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1566579634464752,
|
| 90 |
+
"grad_norm": 38.422706604003906,
|
| 91 |
+
"learning_rate": 4.738903394255875e-05,
|
| 92 |
+
"loss": 2.5929,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.16971279373368145,
|
| 97 |
+
"grad_norm": 49.74126052856445,
|
| 98 |
+
"learning_rate": 4.7171453437771976e-05,
|
| 99 |
+
"loss": 2.5271,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.18276762402088773,
|
| 104 |
+
"grad_norm": 31.077625274658203,
|
| 105 |
+
"learning_rate": 4.6953872932985203e-05,
|
| 106 |
+
"loss": 2.6357,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.195822454308094,
|
| 111 |
+
"grad_norm": 24.155317306518555,
|
| 112 |
+
"learning_rate": 4.673629242819844e-05,
|
| 113 |
+
"loss": 2.5678,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.20887728459530025,
|
| 118 |
+
"grad_norm": 30.788408279418945,
|
| 119 |
+
"learning_rate": 4.651871192341166e-05,
|
| 120 |
+
"loss": 2.8182,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.22193211488250653,
|
| 125 |
+
"grad_norm": 25.434738159179688,
|
| 126 |
+
"learning_rate": 4.630113141862489e-05,
|
| 127 |
+
"loss": 3.303,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.2349869451697128,
|
| 132 |
+
"grad_norm": 37.103668212890625,
|
| 133 |
+
"learning_rate": 4.608355091383813e-05,
|
| 134 |
+
"loss": 2.5012,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.24804177545691905,
|
| 139 |
+
"grad_norm": 70.19502258300781,
|
| 140 |
+
"learning_rate": 4.586597040905135e-05,
|
| 141 |
+
"loss": 2.812,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.26109660574412535,
|
| 146 |
+
"grad_norm": 37.93436050415039,
|
| 147 |
+
"learning_rate": 4.564838990426458e-05,
|
| 148 |
+
"loss": 2.6683,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2741514360313316,
|
| 153 |
+
"grad_norm": 149.83016967773438,
|
| 154 |
+
"learning_rate": 4.543080939947781e-05,
|
| 155 |
+
"loss": 2.9776,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.28720626631853785,
|
| 160 |
+
"grad_norm": 31.74551010131836,
|
| 161 |
+
"learning_rate": 4.521322889469104e-05,
|
| 162 |
+
"loss": 2.9663,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.3002610966057441,
|
| 167 |
+
"grad_norm": 34.0869255065918,
|
| 168 |
+
"learning_rate": 4.4995648389904266e-05,
|
| 169 |
+
"loss": 2.712,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3133159268929504,
|
| 174 |
+
"grad_norm": 28.85022735595703,
|
| 175 |
+
"learning_rate": 4.47780678851175e-05,
|
| 176 |
+
"loss": 2.3012,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.3263707571801567,
|
| 181 |
+
"grad_norm": 36.18962860107422,
|
| 182 |
+
"learning_rate": 4.456048738033072e-05,
|
| 183 |
+
"loss": 2.4339,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.3394255874673629,
|
| 188 |
+
"grad_norm": 27.252077102661133,
|
| 189 |
+
"learning_rate": 4.4342906875543956e-05,
|
| 190 |
+
"loss": 2.851,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.3524804177545692,
|
| 195 |
+
"grad_norm": 38.39606857299805,
|
| 196 |
+
"learning_rate": 4.4125326370757184e-05,
|
| 197 |
+
"loss": 2.6867,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.36553524804177545,
|
| 202 |
+
"grad_norm": 25.8907527923584,
|
| 203 |
+
"learning_rate": 4.390774586597041e-05,
|
| 204 |
+
"loss": 2.4467,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.3785900783289817,
|
| 209 |
+
"grad_norm": 24.98986053466797,
|
| 210 |
+
"learning_rate": 4.369016536118364e-05,
|
| 211 |
+
"loss": 3.0278,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.391644908616188,
|
| 216 |
+
"grad_norm": 24.536916732788086,
|
| 217 |
+
"learning_rate": 4.347258485639687e-05,
|
| 218 |
+
"loss": 2.3857,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.4046997389033943,
|
| 223 |
+
"grad_norm": 22.012798309326172,
|
| 224 |
+
"learning_rate": 4.3255004351610094e-05,
|
| 225 |
+
"loss": 2.1414,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.4177545691906005,
|
| 230 |
+
"grad_norm": 25.466167449951172,
|
| 231 |
+
"learning_rate": 4.303742384682333e-05,
|
| 232 |
+
"loss": 2.5714,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.4308093994778068,
|
| 237 |
+
"grad_norm": 31.785062789916992,
|
| 238 |
+
"learning_rate": 4.281984334203655e-05,
|
| 239 |
+
"loss": 2.4608,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.44386422976501305,
|
| 244 |
+
"grad_norm": 36.67721176147461,
|
| 245 |
+
"learning_rate": 4.2602262837249784e-05,
|
| 246 |
+
"loss": 2.787,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.45691906005221933,
|
| 251 |
+
"grad_norm": 49.02054214477539,
|
| 252 |
+
"learning_rate": 4.238468233246302e-05,
|
| 253 |
+
"loss": 2.7207,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.4699738903394256,
|
| 258 |
+
"grad_norm": 26.876636505126953,
|
| 259 |
+
"learning_rate": 4.216710182767624e-05,
|
| 260 |
+
"loss": 2.6041,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.4830287206266319,
|
| 265 |
+
"grad_norm": 34.956336975097656,
|
| 266 |
+
"learning_rate": 4.1949521322889474e-05,
|
| 267 |
+
"loss": 3.1905,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.4960835509138381,
|
| 272 |
+
"grad_norm": 35.72273254394531,
|
| 273 |
+
"learning_rate": 4.17319408181027e-05,
|
| 274 |
+
"loss": 2.121,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.5091383812010444,
|
| 279 |
+
"grad_norm": 28.895980834960938,
|
| 280 |
+
"learning_rate": 4.151436031331593e-05,
|
| 281 |
+
"loss": 2.6956,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.5221932114882507,
|
| 286 |
+
"grad_norm": 28.925390243530273,
|
| 287 |
+
"learning_rate": 4.129677980852916e-05,
|
| 288 |
+
"loss": 3.0316,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.5352480417754569,
|
| 293 |
+
"grad_norm": 34.79185485839844,
|
| 294 |
+
"learning_rate": 4.107919930374239e-05,
|
| 295 |
+
"loss": 3.0389,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5483028720626631,
|
| 300 |
+
"grad_norm": 30.246923446655273,
|
| 301 |
+
"learning_rate": 4.086161879895561e-05,
|
| 302 |
+
"loss": 2.8086,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.5613577023498695,
|
| 307 |
+
"grad_norm": 32.78372573852539,
|
| 308 |
+
"learning_rate": 4.064403829416885e-05,
|
| 309 |
+
"loss": 2.814,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5744125326370757,
|
| 314 |
+
"grad_norm": 24.346147537231445,
|
| 315 |
+
"learning_rate": 4.0426457789382075e-05,
|
| 316 |
+
"loss": 2.8039,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.587467362924282,
|
| 321 |
+
"grad_norm": 31.166654586791992,
|
| 322 |
+
"learning_rate": 4.02088772845953e-05,
|
| 323 |
+
"loss": 2.9612,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6005221932114883,
|
| 328 |
+
"grad_norm": 23.0938777923584,
|
| 329 |
+
"learning_rate": 3.999129677980853e-05,
|
| 330 |
+
"loss": 2.5463,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6135770234986945,
|
| 335 |
+
"grad_norm": 26.590911865234375,
|
| 336 |
+
"learning_rate": 3.977371627502176e-05,
|
| 337 |
+
"loss": 2.3543,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6266318537859008,
|
| 342 |
+
"grad_norm": 29.803422927856445,
|
| 343 |
+
"learning_rate": 3.9556135770234985e-05,
|
| 344 |
+
"loss": 2.7445,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.639686684073107,
|
| 349 |
+
"grad_norm": 46.66853713989258,
|
| 350 |
+
"learning_rate": 3.933855526544822e-05,
|
| 351 |
+
"loss": 2.4905,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.6527415143603134,
|
| 356 |
+
"grad_norm": 39.04319381713867,
|
| 357 |
+
"learning_rate": 3.912097476066145e-05,
|
| 358 |
+
"loss": 1.8311,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.6657963446475196,
|
| 363 |
+
"grad_norm": 30.50276756286621,
|
| 364 |
+
"learning_rate": 3.8903394255874675e-05,
|
| 365 |
+
"loss": 2.8152,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.6788511749347258,
|
| 370 |
+
"grad_norm": 37.25984191894531,
|
| 371 |
+
"learning_rate": 3.868581375108791e-05,
|
| 372 |
+
"loss": 2.4759,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.6919060052219321,
|
| 377 |
+
"grad_norm": 25.89512062072754,
|
| 378 |
+
"learning_rate": 3.846823324630113e-05,
|
| 379 |
+
"loss": 2.486,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.7049608355091384,
|
| 384 |
+
"grad_norm": 37.318450927734375,
|
| 385 |
+
"learning_rate": 3.8250652741514365e-05,
|
| 386 |
+
"loss": 2.77,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.7180156657963447,
|
| 391 |
+
"grad_norm": 34.45144271850586,
|
| 392 |
+
"learning_rate": 3.803307223672759e-05,
|
| 393 |
+
"loss": 2.2646,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.7310704960835509,
|
| 398 |
+
"grad_norm": 50.494144439697266,
|
| 399 |
+
"learning_rate": 3.781549173194082e-05,
|
| 400 |
+
"loss": 2.7686,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.7441253263707572,
|
| 405 |
+
"grad_norm": 29.753643035888672,
|
| 406 |
+
"learning_rate": 3.759791122715405e-05,
|
| 407 |
+
"loss": 2.6239,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7571801566579635,
|
| 412 |
+
"grad_norm": 39.54145431518555,
|
| 413 |
+
"learning_rate": 3.738033072236728e-05,
|
| 414 |
+
"loss": 2.4995,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7702349869451697,
|
| 419 |
+
"grad_norm": 36.82713317871094,
|
| 420 |
+
"learning_rate": 3.71627502175805e-05,
|
| 421 |
+
"loss": 3.0274,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.783289817232376,
|
| 426 |
+
"grad_norm": 36.62627410888672,
|
| 427 |
+
"learning_rate": 3.694516971279374e-05,
|
| 428 |
+
"loss": 2.2364,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.7963446475195822,
|
| 433 |
+
"grad_norm": 18.279882431030273,
|
| 434 |
+
"learning_rate": 3.6727589208006965e-05,
|
| 435 |
+
"loss": 2.3391,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.8093994778067886,
|
| 440 |
+
"grad_norm": 23.61455535888672,
|
| 441 |
+
"learning_rate": 3.651000870322019e-05,
|
| 442 |
+
"loss": 2.5222,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.8224543080939948,
|
| 447 |
+
"grad_norm": 32.03522872924805,
|
| 448 |
+
"learning_rate": 3.629242819843342e-05,
|
| 449 |
+
"loss": 2.2871,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.835509138381201,
|
| 454 |
+
"grad_norm": 45.24649429321289,
|
| 455 |
+
"learning_rate": 3.607484769364665e-05,
|
| 456 |
+
"loss": 2.6863,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.8485639686684073,
|
| 461 |
+
"grad_norm": 24.39188575744629,
|
| 462 |
+
"learning_rate": 3.5857267188859876e-05,
|
| 463 |
+
"loss": 2.6426,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.8616187989556136,
|
| 468 |
+
"grad_norm": 21.67547607421875,
|
| 469 |
+
"learning_rate": 3.563968668407311e-05,
|
| 470 |
+
"loss": 2.1157,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.8746736292428199,
|
| 475 |
+
"grad_norm": 24.245168685913086,
|
| 476 |
+
"learning_rate": 3.542210617928634e-05,
|
| 477 |
+
"loss": 2.3781,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.8877284595300261,
|
| 482 |
+
"grad_norm": 27.57684326171875,
|
| 483 |
+
"learning_rate": 3.5204525674499566e-05,
|
| 484 |
+
"loss": 2.7673,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.9007832898172323,
|
| 489 |
+
"grad_norm": 28.42872428894043,
|
| 490 |
+
"learning_rate": 3.49869451697128e-05,
|
| 491 |
+
"loss": 2.4503,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.9138381201044387,
|
| 496 |
+
"grad_norm": 39.387813568115234,
|
| 497 |
+
"learning_rate": 3.476936466492602e-05,
|
| 498 |
+
"loss": 2.1765,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.9268929503916449,
|
| 503 |
+
"grad_norm": 20.197811126708984,
|
| 504 |
+
"learning_rate": 3.4551784160139256e-05,
|
| 505 |
+
"loss": 2.555,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.9399477806788512,
|
| 510 |
+
"grad_norm": 22.066137313842773,
|
| 511 |
+
"learning_rate": 3.4334203655352484e-05,
|
| 512 |
+
"loss": 2.4827,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.9530026109660574,
|
| 517 |
+
"grad_norm": 32.67851638793945,
|
| 518 |
+
"learning_rate": 3.411662315056571e-05,
|
| 519 |
+
"loss": 2.762,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9660574412532638,
|
| 524 |
+
"grad_norm": 28.471988677978516,
|
| 525 |
+
"learning_rate": 3.389904264577894e-05,
|
| 526 |
+
"loss": 2.5872,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.97911227154047,
|
| 531 |
+
"grad_norm": 22.934885025024414,
|
| 532 |
+
"learning_rate": 3.368146214099217e-05,
|
| 533 |
+
"loss": 2.8826,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.9921671018276762,
|
| 538 |
+
"grad_norm": 24.063716888427734,
|
| 539 |
+
"learning_rate": 3.3463881636205394e-05,
|
| 540 |
+
"loss": 2.8909,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.0,
|
| 545 |
+
"eval_loss": 2.3283705711364746,
|
| 546 |
+
"eval_runtime": 12.1219,
|
| 547 |
+
"eval_samples_per_second": 112.276,
|
| 548 |
+
"eval_steps_per_second": 14.107,
|
| 549 |
+
"step": 766
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.0052219321148825,
|
| 553 |
+
"grad_norm": 36.15023422241211,
|
| 554 |
+
"learning_rate": 3.324630113141863e-05,
|
| 555 |
+
"loss": 2.5282,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.0182767624020888,
|
| 560 |
+
"grad_norm": 35.99642562866211,
|
| 561 |
+
"learning_rate": 3.3028720626631856e-05,
|
| 562 |
+
"loss": 2.176,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.031331592689295,
|
| 567 |
+
"grad_norm": 30.217031478881836,
|
| 568 |
+
"learning_rate": 3.2811140121845084e-05,
|
| 569 |
+
"loss": 2.2727,
|
| 570 |
+
"step": 790
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.0443864229765012,
|
| 574 |
+
"grad_norm": 29.16168212890625,
|
| 575 |
+
"learning_rate": 3.259355961705831e-05,
|
| 576 |
+
"loss": 2.0302,
|
| 577 |
+
"step": 800
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.0574412532637076,
|
| 581 |
+
"grad_norm": 25.400541305541992,
|
| 582 |
+
"learning_rate": 3.237597911227154e-05,
|
| 583 |
+
"loss": 2.3489,
|
| 584 |
+
"step": 810
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.0704960835509139,
|
| 588 |
+
"grad_norm": 21.281591415405273,
|
| 589 |
+
"learning_rate": 3.215839860748477e-05,
|
| 590 |
+
"loss": 2.3976,
|
| 591 |
+
"step": 820
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.08355091383812,
|
| 595 |
+
"grad_norm": 23.941238403320312,
|
| 596 |
+
"learning_rate": 3.1940818102698e-05,
|
| 597 |
+
"loss": 2.6169,
|
| 598 |
+
"step": 830
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.0966057441253263,
|
| 602 |
+
"grad_norm": 26.626665115356445,
|
| 603 |
+
"learning_rate": 3.172323759791123e-05,
|
| 604 |
+
"loss": 2.6102,
|
| 605 |
+
"step": 840
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.1096605744125327,
|
| 609 |
+
"grad_norm": 28.539621353149414,
|
| 610 |
+
"learning_rate": 3.150565709312446e-05,
|
| 611 |
+
"loss": 2.019,
|
| 612 |
+
"step": 850
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.122715404699739,
|
| 616 |
+
"grad_norm": 36.77280044555664,
|
| 617 |
+
"learning_rate": 3.128807658833769e-05,
|
| 618 |
+
"loss": 2.2338,
|
| 619 |
+
"step": 860
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.1357702349869452,
|
| 623 |
+
"grad_norm": 38.7175407409668,
|
| 624 |
+
"learning_rate": 3.107049608355091e-05,
|
| 625 |
+
"loss": 2.7259,
|
| 626 |
+
"step": 870
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.1488250652741514,
|
| 630 |
+
"grad_norm": 32.9740104675293,
|
| 631 |
+
"learning_rate": 3.085291557876415e-05,
|
| 632 |
+
"loss": 2.294,
|
| 633 |
+
"step": 880
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.1618798955613576,
|
| 637 |
+
"grad_norm": 35.01115036010742,
|
| 638 |
+
"learning_rate": 3.0635335073977374e-05,
|
| 639 |
+
"loss": 2.5392,
|
| 640 |
+
"step": 890
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.174934725848564,
|
| 644 |
+
"grad_norm": 40.960968017578125,
|
| 645 |
+
"learning_rate": 3.0417754569190606e-05,
|
| 646 |
+
"loss": 1.8062,
|
| 647 |
+
"step": 900
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.1879895561357703,
|
| 651 |
+
"grad_norm": 24.627063751220703,
|
| 652 |
+
"learning_rate": 3.020017406440383e-05,
|
| 653 |
+
"loss": 2.746,
|
| 654 |
+
"step": 910
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.2010443864229765,
|
| 658 |
+
"grad_norm": 28.861692428588867,
|
| 659 |
+
"learning_rate": 2.998259355961706e-05,
|
| 660 |
+
"loss": 2.1434,
|
| 661 |
+
"step": 920
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.2140992167101827,
|
| 665 |
+
"grad_norm": 30.32466697692871,
|
| 666 |
+
"learning_rate": 2.976501305483029e-05,
|
| 667 |
+
"loss": 2.8282,
|
| 668 |
+
"step": 930
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.227154046997389,
|
| 672 |
+
"grad_norm": 45.79476547241211,
|
| 673 |
+
"learning_rate": 2.954743255004352e-05,
|
| 674 |
+
"loss": 2.1972,
|
| 675 |
+
"step": 940
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.2402088772845954,
|
| 679 |
+
"grad_norm": 34.27708435058594,
|
| 680 |
+
"learning_rate": 2.9329852045256744e-05,
|
| 681 |
+
"loss": 3.0359,
|
| 682 |
+
"step": 950
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.2532637075718016,
|
| 686 |
+
"grad_norm": 33.62773513793945,
|
| 687 |
+
"learning_rate": 2.9112271540469975e-05,
|
| 688 |
+
"loss": 2.4254,
|
| 689 |
+
"step": 960
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.2663185378590078,
|
| 693 |
+
"grad_norm": 26.693920135498047,
|
| 694 |
+
"learning_rate": 2.8894691035683203e-05,
|
| 695 |
+
"loss": 1.883,
|
| 696 |
+
"step": 970
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.279373368146214,
|
| 700 |
+
"grad_norm": 36.45111083984375,
|
| 701 |
+
"learning_rate": 2.8677110530896434e-05,
|
| 702 |
+
"loss": 2.232,
|
| 703 |
+
"step": 980
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.2924281984334205,
|
| 707 |
+
"grad_norm": 34.38032913208008,
|
| 708 |
+
"learning_rate": 2.845953002610966e-05,
|
| 709 |
+
"loss": 2.4289,
|
| 710 |
+
"step": 990
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.3054830287206267,
|
| 714 |
+
"grad_norm": 19.068925857543945,
|
| 715 |
+
"learning_rate": 2.8241949521322892e-05,
|
| 716 |
+
"loss": 2.473,
|
| 717 |
+
"step": 1000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.318537859007833,
|
| 721 |
+
"grad_norm": 30.37474250793457,
|
| 722 |
+
"learning_rate": 2.8024369016536117e-05,
|
| 723 |
+
"loss": 2.6123,
|
| 724 |
+
"step": 1010
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.3315926892950392,
|
| 728 |
+
"grad_norm": 29.176149368286133,
|
| 729 |
+
"learning_rate": 2.7806788511749348e-05,
|
| 730 |
+
"loss": 2.3158,
|
| 731 |
+
"step": 1020
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.3446475195822454,
|
| 735 |
+
"grad_norm": 27.721553802490234,
|
| 736 |
+
"learning_rate": 2.758920800696258e-05,
|
| 737 |
+
"loss": 2.3646,
|
| 738 |
+
"step": 1030
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.3577023498694518,
|
| 742 |
+
"grad_norm": 12.712676048278809,
|
| 743 |
+
"learning_rate": 2.7371627502175807e-05,
|
| 744 |
+
"loss": 2.1114,
|
| 745 |
+
"step": 1040
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.370757180156658,
|
| 749 |
+
"grad_norm": 23.869230270385742,
|
| 750 |
+
"learning_rate": 2.7154046997389038e-05,
|
| 751 |
+
"loss": 2.1335,
|
| 752 |
+
"step": 1050
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.3838120104438643,
|
| 756 |
+
"grad_norm": 23.66261100769043,
|
| 757 |
+
"learning_rate": 2.6936466492602262e-05,
|
| 758 |
+
"loss": 2.0774,
|
| 759 |
+
"step": 1060
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.3968668407310705,
|
| 763 |
+
"grad_norm": 44.03451156616211,
|
| 764 |
+
"learning_rate": 2.6718885987815496e-05,
|
| 765 |
+
"loss": 2.6329,
|
| 766 |
+
"step": 1070
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.4099216710182767,
|
| 770 |
+
"grad_norm": 20.849573135375977,
|
| 771 |
+
"learning_rate": 2.650130548302872e-05,
|
| 772 |
+
"loss": 2.8549,
|
| 773 |
+
"step": 1080
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.4229765013054831,
|
| 777 |
+
"grad_norm": 27.392250061035156,
|
| 778 |
+
"learning_rate": 2.6283724978241952e-05,
|
| 779 |
+
"loss": 2.4537,
|
| 780 |
+
"step": 1090
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.4360313315926894,
|
| 784 |
+
"grad_norm": 25.04733657836914,
|
| 785 |
+
"learning_rate": 2.606614447345518e-05,
|
| 786 |
+
"loss": 1.9425,
|
| 787 |
+
"step": 1100
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.4490861618798956,
|
| 791 |
+
"grad_norm": 32.247894287109375,
|
| 792 |
+
"learning_rate": 2.584856396866841e-05,
|
| 793 |
+
"loss": 2.0056,
|
| 794 |
+
"step": 1110
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.4621409921671018,
|
| 798 |
+
"grad_norm": 32.498191833496094,
|
| 799 |
+
"learning_rate": 2.5630983463881635e-05,
|
| 800 |
+
"loss": 2.6457,
|
| 801 |
+
"step": 1120
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.475195822454308,
|
| 805 |
+
"grad_norm": 26.728214263916016,
|
| 806 |
+
"learning_rate": 2.5413402959094866e-05,
|
| 807 |
+
"loss": 1.6954,
|
| 808 |
+
"step": 1130
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.4882506527415145,
|
| 812 |
+
"grad_norm": 63.32724380493164,
|
| 813 |
+
"learning_rate": 2.5195822454308094e-05,
|
| 814 |
+
"loss": 2.5008,
|
| 815 |
+
"step": 1140
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.5013054830287205,
|
| 819 |
+
"grad_norm": 24.65690803527832,
|
| 820 |
+
"learning_rate": 2.4978241949521325e-05,
|
| 821 |
+
"loss": 1.5901,
|
| 822 |
+
"step": 1150
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.514360313315927,
|
| 826 |
+
"grad_norm": 19.37055206298828,
|
| 827 |
+
"learning_rate": 2.4760661444734552e-05,
|
| 828 |
+
"loss": 1.9548,
|
| 829 |
+
"step": 1160
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.5274151436031331,
|
| 833 |
+
"grad_norm": 23.001806259155273,
|
| 834 |
+
"learning_rate": 2.4543080939947783e-05,
|
| 835 |
+
"loss": 2.382,
|
| 836 |
+
"step": 1170
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.5404699738903394,
|
| 840 |
+
"grad_norm": 26.997100830078125,
|
| 841 |
+
"learning_rate": 2.432550043516101e-05,
|
| 842 |
+
"loss": 2.2834,
|
| 843 |
+
"step": 1180
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.5535248041775458,
|
| 847 |
+
"grad_norm": 22.0489559173584,
|
| 848 |
+
"learning_rate": 2.410791993037424e-05,
|
| 849 |
+
"loss": 2.1768,
|
| 850 |
+
"step": 1190
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.566579634464752,
|
| 854 |
+
"grad_norm": 29.986967086791992,
|
| 855 |
+
"learning_rate": 2.389033942558747e-05,
|
| 856 |
+
"loss": 2.0854,
|
| 857 |
+
"step": 1200
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.5796344647519582,
|
| 861 |
+
"grad_norm": 34.0990104675293,
|
| 862 |
+
"learning_rate": 2.3672758920800698e-05,
|
| 863 |
+
"loss": 2.4429,
|
| 864 |
+
"step": 1210
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.5926892950391645,
|
| 868 |
+
"grad_norm": 25.31661033630371,
|
| 869 |
+
"learning_rate": 2.3455178416013925e-05,
|
| 870 |
+
"loss": 2.3145,
|
| 871 |
+
"step": 1220
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.6057441253263707,
|
| 875 |
+
"grad_norm": 30.300716400146484,
|
| 876 |
+
"learning_rate": 2.3237597911227156e-05,
|
| 877 |
+
"loss": 1.9357,
|
| 878 |
+
"step": 1230
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.6187989556135771,
|
| 882 |
+
"grad_norm": 25.158327102661133,
|
| 883 |
+
"learning_rate": 2.3020017406440384e-05,
|
| 884 |
+
"loss": 2.3195,
|
| 885 |
+
"step": 1240
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.6318537859007833,
|
| 889 |
+
"grad_norm": 33.35712432861328,
|
| 890 |
+
"learning_rate": 2.280243690165361e-05,
|
| 891 |
+
"loss": 2.4724,
|
| 892 |
+
"step": 1250
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.6449086161879896,
|
| 896 |
+
"grad_norm": 22.938852310180664,
|
| 897 |
+
"learning_rate": 2.258485639686684e-05,
|
| 898 |
+
"loss": 2.3876,
|
| 899 |
+
"step": 1260
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.6579634464751958,
|
| 903 |
+
"grad_norm": 26.44889259338379,
|
| 904 |
+
"learning_rate": 2.236727589208007e-05,
|
| 905 |
+
"loss": 2.1164,
|
| 906 |
+
"step": 1270
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.671018276762402,
|
| 910 |
+
"grad_norm": 28.251296997070312,
|
| 911 |
+
"learning_rate": 2.2149695387293298e-05,
|
| 912 |
+
"loss": 2.3045,
|
| 913 |
+
"step": 1280
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.6840731070496084,
|
| 917 |
+
"grad_norm": 28.00015640258789,
|
| 918 |
+
"learning_rate": 2.193211488250653e-05,
|
| 919 |
+
"loss": 1.9546,
|
| 920 |
+
"step": 1290
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.6971279373368147,
|
| 924 |
+
"grad_norm": 20.0263729095459,
|
| 925 |
+
"learning_rate": 2.171453437771976e-05,
|
| 926 |
+
"loss": 2.2491,
|
| 927 |
+
"step": 1300
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.7101827676240209,
|
| 931 |
+
"grad_norm": 23.335580825805664,
|
| 932 |
+
"learning_rate": 2.1496953872932988e-05,
|
| 933 |
+
"loss": 2.5543,
|
| 934 |
+
"step": 1310
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.723237597911227,
|
| 938 |
+
"grad_norm": 25.97711181640625,
|
| 939 |
+
"learning_rate": 2.1279373368146216e-05,
|
| 940 |
+
"loss": 2.2486,
|
| 941 |
+
"step": 1320
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.7362924281984333,
|
| 945 |
+
"grad_norm": 24.391855239868164,
|
| 946 |
+
"learning_rate": 2.1061792863359443e-05,
|
| 947 |
+
"loss": 2.3483,
|
| 948 |
+
"step": 1330
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.7493472584856398,
|
| 952 |
+
"grad_norm": 29.249792098999023,
|
| 953 |
+
"learning_rate": 2.0844212358572674e-05,
|
| 954 |
+
"loss": 2.3703,
|
| 955 |
+
"step": 1340
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.762402088772846,
|
| 959 |
+
"grad_norm": 20.213987350463867,
|
| 960 |
+
"learning_rate": 2.0626631853785902e-05,
|
| 961 |
+
"loss": 2.3178,
|
| 962 |
+
"step": 1350
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.7754569190600522,
|
| 966 |
+
"grad_norm": 33.050018310546875,
|
| 967 |
+
"learning_rate": 2.040905134899913e-05,
|
| 968 |
+
"loss": 2.2622,
|
| 969 |
+
"step": 1360
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.7885117493472587,
|
| 973 |
+
"grad_norm": 25.058115005493164,
|
| 974 |
+
"learning_rate": 2.019147084421236e-05,
|
| 975 |
+
"loss": 2.2577,
|
| 976 |
+
"step": 1370
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.8015665796344646,
|
| 980 |
+
"grad_norm": 34.79226303100586,
|
| 981 |
+
"learning_rate": 1.997389033942559e-05,
|
| 982 |
+
"loss": 2.1586,
|
| 983 |
+
"step": 1380
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.814621409921671,
|
| 987 |
+
"grad_norm": 38.56571578979492,
|
| 988 |
+
"learning_rate": 1.9756309834638816e-05,
|
| 989 |
+
"loss": 2.0911,
|
| 990 |
+
"step": 1390
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.8276762402088773,
|
| 994 |
+
"grad_norm": 27.778825759887695,
|
| 995 |
+
"learning_rate": 1.9538729329852047e-05,
|
| 996 |
+
"loss": 2.2379,
|
| 997 |
+
"step": 1400
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.8407310704960835,
|
| 1001 |
+
"grad_norm": 27.160274505615234,
|
| 1002 |
+
"learning_rate": 1.9321148825065275e-05,
|
| 1003 |
+
"loss": 2.0847,
|
| 1004 |
+
"step": 1410
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.85378590078329,
|
| 1008 |
+
"grad_norm": 26.11197853088379,
|
| 1009 |
+
"learning_rate": 1.9103568320278503e-05,
|
| 1010 |
+
"loss": 2.7021,
|
| 1011 |
+
"step": 1420
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.866840731070496,
|
| 1015 |
+
"grad_norm": 28.448244094848633,
|
| 1016 |
+
"learning_rate": 1.8885987815491734e-05,
|
| 1017 |
+
"loss": 2.1596,
|
| 1018 |
+
"step": 1430
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.8798955613577024,
|
| 1022 |
+
"grad_norm": 49.074729919433594,
|
| 1023 |
+
"learning_rate": 1.866840731070496e-05,
|
| 1024 |
+
"loss": 2.5753,
|
| 1025 |
+
"step": 1440
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.8929503916449086,
|
| 1029 |
+
"grad_norm": 21.96980094909668,
|
| 1030 |
+
"learning_rate": 1.845082680591819e-05,
|
| 1031 |
+
"loss": 2.5118,
|
| 1032 |
+
"step": 1450
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.9060052219321149,
|
| 1036 |
+
"grad_norm": 20.993181228637695,
|
| 1037 |
+
"learning_rate": 1.823324630113142e-05,
|
| 1038 |
+
"loss": 2.2332,
|
| 1039 |
+
"step": 1460
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.9190600522193213,
|
| 1043 |
+
"grad_norm": 20.049209594726562,
|
| 1044 |
+
"learning_rate": 1.801566579634465e-05,
|
| 1045 |
+
"loss": 2.1748,
|
| 1046 |
+
"step": 1470
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.9321148825065273,
|
| 1050 |
+
"grad_norm": 35.51521682739258,
|
| 1051 |
+
"learning_rate": 1.779808529155788e-05,
|
| 1052 |
+
"loss": 2.781,
|
| 1053 |
+
"step": 1480
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.9451697127937337,
|
| 1057 |
+
"grad_norm": 25.36643409729004,
|
| 1058 |
+
"learning_rate": 1.7580504786771106e-05,
|
| 1059 |
+
"loss": 2.7489,
|
| 1060 |
+
"step": 1490
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.95822454308094,
|
| 1064 |
+
"grad_norm": 38.09309387207031,
|
| 1065 |
+
"learning_rate": 1.7362924281984334e-05,
|
| 1066 |
+
"loss": 2.1326,
|
| 1067 |
+
"step": 1500
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.9712793733681462,
|
| 1071 |
+
"grad_norm": 30.636632919311523,
|
| 1072 |
+
"learning_rate": 1.7145343777197565e-05,
|
| 1073 |
+
"loss": 2.2151,
|
| 1074 |
+
"step": 1510
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.9843342036553526,
|
| 1078 |
+
"grad_norm": 27.038352966308594,
|
| 1079 |
+
"learning_rate": 1.6927763272410793e-05,
|
| 1080 |
+
"loss": 2.4523,
|
| 1081 |
+
"step": 1520
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.9973890339425586,
|
| 1085 |
+
"grad_norm": 19.101573944091797,
|
| 1086 |
+
"learning_rate": 1.671018276762402e-05,
|
| 1087 |
+
"loss": 2.518,
|
| 1088 |
+
"step": 1530
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 2.0,
|
| 1092 |
+
"eval_loss": NaN,
|
| 1093 |
+
"eval_runtime": 11.9494,
|
| 1094 |
+
"eval_samples_per_second": 113.897,
|
| 1095 |
+
"eval_steps_per_second": 14.31,
|
| 1096 |
+
"step": 1532
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 2.010443864229765,
|
| 1100 |
+
"grad_norm": 28.463035583496094,
|
| 1101 |
+
"learning_rate": 1.649260226283725e-05,
|
| 1102 |
+
"loss": 1.8945,
|
| 1103 |
+
"step": 1540
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 2.023498694516971,
|
| 1107 |
+
"grad_norm": 30.520097732543945,
|
| 1108 |
+
"learning_rate": 1.627502175805048e-05,
|
| 1109 |
+
"loss": 2.3685,
|
| 1110 |
+
"step": 1550
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 2.0365535248041775,
|
| 1114 |
+
"grad_norm": 19.876482009887695,
|
| 1115 |
+
"learning_rate": 1.6057441253263707e-05,
|
| 1116 |
+
"loss": 2.1494,
|
| 1117 |
+
"step": 1560
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 2.049608355091384,
|
| 1121 |
+
"grad_norm": 23.423219680786133,
|
| 1122 |
+
"learning_rate": 1.5839860748476938e-05,
|
| 1123 |
+
"loss": 1.9791,
|
| 1124 |
+
"step": 1570
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 2.06266318537859,
|
| 1128 |
+
"grad_norm": 20.257450103759766,
|
| 1129 |
+
"learning_rate": 1.5622280243690166e-05,
|
| 1130 |
+
"loss": 2.2526,
|
| 1131 |
+
"step": 1580
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 2.0757180156657964,
|
| 1135 |
+
"grad_norm": 47.68708038330078,
|
| 1136 |
+
"learning_rate": 1.5404699738903393e-05,
|
| 1137 |
+
"loss": 1.9962,
|
| 1138 |
+
"step": 1590
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 2.0887728459530024,
|
| 1142 |
+
"grad_norm": 27.561660766601562,
|
| 1143 |
+
"learning_rate": 1.5187119234116623e-05,
|
| 1144 |
+
"loss": 1.6106,
|
| 1145 |
+
"step": 1600
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 2.101827676240209,
|
| 1149 |
+
"grad_norm": 26.832944869995117,
|
| 1150 |
+
"learning_rate": 1.4969538729329852e-05,
|
| 1151 |
+
"loss": 1.5015,
|
| 1152 |
+
"step": 1610
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 2.1148825065274153,
|
| 1156 |
+
"grad_norm": 45.05983352661133,
|
| 1157 |
+
"learning_rate": 1.475195822454308e-05,
|
| 1158 |
+
"loss": 1.8897,
|
| 1159 |
+
"step": 1620
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 2.1279373368146213,
|
| 1163 |
+
"grad_norm": 19.24533462524414,
|
| 1164 |
+
"learning_rate": 1.4534377719756313e-05,
|
| 1165 |
+
"loss": 2.4187,
|
| 1166 |
+
"step": 1630
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 2.1409921671018277,
|
| 1170 |
+
"grad_norm": 29.476770401000977,
|
| 1171 |
+
"learning_rate": 1.431679721496954e-05,
|
| 1172 |
+
"loss": 2.1733,
|
| 1173 |
+
"step": 1640
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 2.1540469973890337,
|
| 1177 |
+
"grad_norm": 26.505355834960938,
|
| 1178 |
+
"learning_rate": 1.409921671018277e-05,
|
| 1179 |
+
"loss": 1.8432,
|
| 1180 |
+
"step": 1650
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 2.16710182767624,
|
| 1184 |
+
"grad_norm": 28.1693058013916,
|
| 1185 |
+
"learning_rate": 1.3881636205395997e-05,
|
| 1186 |
+
"loss": 2.1419,
|
| 1187 |
+
"step": 1660
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 2.1801566579634466,
|
| 1191 |
+
"grad_norm": 37.704498291015625,
|
| 1192 |
+
"learning_rate": 1.3664055700609227e-05,
|
| 1193 |
+
"loss": 2.7488,
|
| 1194 |
+
"step": 1670
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 2.1932114882506526,
|
| 1198 |
+
"grad_norm": 22.38772964477539,
|
| 1199 |
+
"learning_rate": 1.3446475195822456e-05,
|
| 1200 |
+
"loss": 2.3531,
|
| 1201 |
+
"step": 1680
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 2.206266318537859,
|
| 1205 |
+
"grad_norm": 22.58838653564453,
|
| 1206 |
+
"learning_rate": 1.3228894691035684e-05,
|
| 1207 |
+
"loss": 1.8415,
|
| 1208 |
+
"step": 1690
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 2.2193211488250655,
|
| 1212 |
+
"grad_norm": 30.01149559020996,
|
| 1213 |
+
"learning_rate": 1.3011314186248913e-05,
|
| 1214 |
+
"loss": 2.0757,
|
| 1215 |
+
"step": 1700
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 2.2323759791122715,
|
| 1219 |
+
"grad_norm": 23.964759826660156,
|
| 1220 |
+
"learning_rate": 1.2793733681462141e-05,
|
| 1221 |
+
"loss": 2.4956,
|
| 1222 |
+
"step": 1710
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 2.245430809399478,
|
| 1226 |
+
"grad_norm": 33.133541107177734,
|
| 1227 |
+
"learning_rate": 1.257615317667537e-05,
|
| 1228 |
+
"loss": 1.971,
|
| 1229 |
+
"step": 1720
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 2.258485639686684,
|
| 1233 |
+
"grad_norm": 27.34188461303711,
|
| 1234 |
+
"learning_rate": 1.23585726718886e-05,
|
| 1235 |
+
"loss": 2.2084,
|
| 1236 |
+
"step": 1730
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 2.2715404699738904,
|
| 1240 |
+
"grad_norm": 25.62513542175293,
|
| 1241 |
+
"learning_rate": 1.2140992167101827e-05,
|
| 1242 |
+
"loss": 1.9445,
|
| 1243 |
+
"step": 1740
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 2.2845953002610964,
|
| 1247 |
+
"grad_norm": 33.618385314941406,
|
| 1248 |
+
"learning_rate": 1.1923411662315057e-05,
|
| 1249 |
+
"loss": 2.4505,
|
| 1250 |
+
"step": 1750
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 2.297650130548303,
|
| 1254 |
+
"grad_norm": 25.787757873535156,
|
| 1255 |
+
"learning_rate": 1.1705831157528286e-05,
|
| 1256 |
+
"loss": 2.748,
|
| 1257 |
+
"step": 1760
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 2.3107049608355092,
|
| 1261 |
+
"grad_norm": 24.533018112182617,
|
| 1262 |
+
"learning_rate": 1.1488250652741515e-05,
|
| 1263 |
+
"loss": 1.496,
|
| 1264 |
+
"step": 1770
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 2.3237597911227152,
|
| 1268 |
+
"grad_norm": 41.41205596923828,
|
| 1269 |
+
"learning_rate": 1.1270670147954745e-05,
|
| 1270 |
+
"loss": 2.5213,
|
| 1271 |
+
"step": 1780
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 2.3368146214099217,
|
| 1275 |
+
"grad_norm": 20.163238525390625,
|
| 1276 |
+
"learning_rate": 1.1053089643167972e-05,
|
| 1277 |
+
"loss": 1.8429,
|
| 1278 |
+
"step": 1790
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 2.349869451697128,
|
| 1282 |
+
"grad_norm": 36.58127975463867,
|
| 1283 |
+
"learning_rate": 1.0835509138381202e-05,
|
| 1284 |
+
"loss": 2.617,
|
| 1285 |
+
"step": 1800
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 2.362924281984334,
|
| 1289 |
+
"grad_norm": 29.932636260986328,
|
| 1290 |
+
"learning_rate": 1.061792863359443e-05,
|
| 1291 |
+
"loss": 2.2462,
|
| 1292 |
+
"step": 1810
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 2.3759791122715406,
|
| 1296 |
+
"grad_norm": 20.780025482177734,
|
| 1297 |
+
"learning_rate": 1.0400348128807659e-05,
|
| 1298 |
+
"loss": 1.7878,
|
| 1299 |
+
"step": 1820
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.3890339425587466,
|
| 1303 |
+
"grad_norm": 26.663557052612305,
|
| 1304 |
+
"learning_rate": 1.0182767624020888e-05,
|
| 1305 |
+
"loss": 2.2757,
|
| 1306 |
+
"step": 1830
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 2.402088772845953,
|
| 1310 |
+
"grad_norm": 34.194435119628906,
|
| 1311 |
+
"learning_rate": 9.965187119234116e-06,
|
| 1312 |
+
"loss": 1.8959,
|
| 1313 |
+
"step": 1840
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 2.4151436031331595,
|
| 1317 |
+
"grad_norm": 13.468405723571777,
|
| 1318 |
+
"learning_rate": 9.747606614447347e-06,
|
| 1319 |
+
"loss": 2.2176,
|
| 1320 |
+
"step": 1850
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 2.4281984334203655,
|
| 1324 |
+
"grad_norm": 36.56803512573242,
|
| 1325 |
+
"learning_rate": 9.530026109660575e-06,
|
| 1326 |
+
"loss": 2.2373,
|
| 1327 |
+
"step": 1860
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 2.441253263707572,
|
| 1331 |
+
"grad_norm": 25.580917358398438,
|
| 1332 |
+
"learning_rate": 9.312445604873804e-06,
|
| 1333 |
+
"loss": 2.004,
|
| 1334 |
+
"step": 1870
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 2.454308093994778,
|
| 1338 |
+
"grad_norm": 47.98051071166992,
|
| 1339 |
+
"learning_rate": 9.094865100087033e-06,
|
| 1340 |
+
"loss": 2.09,
|
| 1341 |
+
"step": 1880
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 2.4673629242819843,
|
| 1345 |
+
"grad_norm": 44.3414421081543,
|
| 1346 |
+
"learning_rate": 8.877284595300261e-06,
|
| 1347 |
+
"loss": 2.4412,
|
| 1348 |
+
"step": 1890
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 2.480417754569191,
|
| 1352 |
+
"grad_norm": 22.451644897460938,
|
| 1353 |
+
"learning_rate": 8.65970409051349e-06,
|
| 1354 |
+
"loss": 1.5646,
|
| 1355 |
+
"step": 1900
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 2.493472584856397,
|
| 1359 |
+
"grad_norm": 21.15165138244629,
|
| 1360 |
+
"learning_rate": 8.442123585726718e-06,
|
| 1361 |
+
"loss": 1.5899,
|
| 1362 |
+
"step": 1910
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 2.506527415143603,
|
| 1366 |
+
"grad_norm": 49.44068145751953,
|
| 1367 |
+
"learning_rate": 8.224543080939948e-06,
|
| 1368 |
+
"loss": 2.0553,
|
| 1369 |
+
"step": 1920
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 2.5195822454308097,
|
| 1373 |
+
"grad_norm": 30.540063858032227,
|
| 1374 |
+
"learning_rate": 8.006962576153177e-06,
|
| 1375 |
+
"loss": 1.7215,
|
| 1376 |
+
"step": 1930
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 2.5326370757180157,
|
| 1380 |
+
"grad_norm": 29.058853149414062,
|
| 1381 |
+
"learning_rate": 7.789382071366406e-06,
|
| 1382 |
+
"loss": 1.7025,
|
| 1383 |
+
"step": 1940
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 2.5456919060052217,
|
| 1387 |
+
"grad_norm": 36.38139343261719,
|
| 1388 |
+
"learning_rate": 7.571801566579635e-06,
|
| 1389 |
+
"loss": 1.9976,
|
| 1390 |
+
"step": 1950
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 2.558746736292428,
|
| 1394 |
+
"grad_norm": 21.742773056030273,
|
| 1395 |
+
"learning_rate": 7.354221061792864e-06,
|
| 1396 |
+
"loss": 1.8528,
|
| 1397 |
+
"step": 1960
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 2.5718015665796345,
|
| 1401 |
+
"grad_norm": 22.30496597290039,
|
| 1402 |
+
"learning_rate": 7.136640557006093e-06,
|
| 1403 |
+
"loss": 1.726,
|
| 1404 |
+
"step": 1970
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.584856396866841,
|
| 1408 |
+
"grad_norm": 28.57794761657715,
|
| 1409 |
+
"learning_rate": 6.919060052219321e-06,
|
| 1410 |
+
"loss": 1.9179,
|
| 1411 |
+
"step": 1980
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 2.597911227154047,
|
| 1415 |
+
"grad_norm": 35.680999755859375,
|
| 1416 |
+
"learning_rate": 6.70147954743255e-06,
|
| 1417 |
+
"loss": 1.9789,
|
| 1418 |
+
"step": 1990
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 2.6109660574412534,
|
| 1422 |
+
"grad_norm": 41.53910827636719,
|
| 1423 |
+
"learning_rate": 6.483899042645779e-06,
|
| 1424 |
+
"loss": 2.3311,
|
| 1425 |
+
"step": 2000
|
| 1426 |
+
}
|
| 1427 |
+
],
|
| 1428 |
+
"logging_steps": 10,
|
| 1429 |
+
"max_steps": 2298,
|
| 1430 |
+
"num_input_tokens_seen": 0,
|
| 1431 |
+
"num_train_epochs": 3,
|
| 1432 |
+
"save_steps": 500,
|
| 1433 |
+
"stateful_callbacks": {
|
| 1434 |
+
"TrainerControl": {
|
| 1435 |
+
"args": {
|
| 1436 |
+
"should_epoch_stop": false,
|
| 1437 |
+
"should_evaluate": false,
|
| 1438 |
+
"should_log": false,
|
| 1439 |
+
"should_save": true,
|
| 1440 |
+
"should_training_stop": false
|
| 1441 |
+
},
|
| 1442 |
+
"attributes": {}
|
| 1443 |
+
}
|
| 1444 |
+
},
|
| 1445 |
+
"total_flos": 2103832360255488.0,
|
| 1446 |
+
"train_batch_size": 16,
|
| 1447 |
+
"trial_name": null,
|
| 1448 |
+
"trial_params": null
|
| 1449 |
+
}
|
checkpoint-2000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f86adb9de168a824dd2efa36c83bb2323ddc512b2622a8e0f1f31147371530a
|
| 3 |
+
size 5304
|
checkpoint-2298/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "dbmdz/bert-base-turkish-uncased",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"torch_dtype": "float32",
|
| 21 |
+
"transformers_version": "4.48.2",
|
| 22 |
+
"type_vocab_size": 2,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 32000
|
| 25 |
+
}
|
checkpoint-2298/generation_config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"pad_token_id": 0,
|
| 4 |
+
"transformers_version": "4.48.2"
|
| 5 |
+
}
|
checkpoint-2298/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f4e3329028953d7cfc643ab3b75ca2c123e19c0e0625c51591d26c864159317
|
| 3 |
+
size 442627224
|
checkpoint-2298/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5ddbcbadbf7846e298c0ee2fafce047cac5fffef236e230ec3e075d530760879
|
| 3 |
+
size 885375994
|
checkpoint-2298/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c17c1858dafa976c969b14eb7d209a50a9b5fc8389b607d66bb2d4a1ef21d657
|
| 3 |
+
size 14244
|
checkpoint-2298/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17bfbea152dd4153588f1e8d24d88690301542b6a3e896e917ebf38eef67b1b0
|
| 3 |
+
size 1064
|
checkpoint-2298/trainer_state.json
ADDED
|
@@ -0,0 +1,1652 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2298,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.013054830287206266,
|
| 13 |
+
"grad_norm": 43.415225982666016,
|
| 14 |
+
"learning_rate": 4.978241949521324e-05,
|
| 15 |
+
"loss": 2.9884,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02610966057441253,
|
| 20 |
+
"grad_norm": 49.437705993652344,
|
| 21 |
+
"learning_rate": 4.956483899042646e-05,
|
| 22 |
+
"loss": 2.9654,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0391644908616188,
|
| 27 |
+
"grad_norm": 47.24225997924805,
|
| 28 |
+
"learning_rate": 4.934725848563969e-05,
|
| 29 |
+
"loss": 3.5102,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.05221932114882506,
|
| 34 |
+
"grad_norm": 32.93499755859375,
|
| 35 |
+
"learning_rate": 4.912967798085292e-05,
|
| 36 |
+
"loss": 2.4143,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.06527415143603134,
|
| 41 |
+
"grad_norm": 31.01405143737793,
|
| 42 |
+
"learning_rate": 4.891209747606615e-05,
|
| 43 |
+
"loss": 2.6864,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0783289817232376,
|
| 48 |
+
"grad_norm": 77.32862854003906,
|
| 49 |
+
"learning_rate": 4.8694516971279375e-05,
|
| 50 |
+
"loss": 2.7403,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.09138381201044386,
|
| 55 |
+
"grad_norm": 28.313295364379883,
|
| 56 |
+
"learning_rate": 4.84769364664926e-05,
|
| 57 |
+
"loss": 2.5107,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.10443864229765012,
|
| 62 |
+
"grad_norm": 38.45579528808594,
|
| 63 |
+
"learning_rate": 4.825935596170583e-05,
|
| 64 |
+
"loss": 2.6546,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1174934725848564,
|
| 69 |
+
"grad_norm": 41.91643142700195,
|
| 70 |
+
"learning_rate": 4.8041775456919065e-05,
|
| 71 |
+
"loss": 2.82,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.13054830287206268,
|
| 76 |
+
"grad_norm": 36.32301712036133,
|
| 77 |
+
"learning_rate": 4.782419495213229e-05,
|
| 78 |
+
"loss": 2.9063,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.14360313315926893,
|
| 83 |
+
"grad_norm": 30.05735969543457,
|
| 84 |
+
"learning_rate": 4.760661444734552e-05,
|
| 85 |
+
"loss": 2.5942,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1566579634464752,
|
| 90 |
+
"grad_norm": 38.422706604003906,
|
| 91 |
+
"learning_rate": 4.738903394255875e-05,
|
| 92 |
+
"loss": 2.5929,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.16971279373368145,
|
| 97 |
+
"grad_norm": 49.74126052856445,
|
| 98 |
+
"learning_rate": 4.7171453437771976e-05,
|
| 99 |
+
"loss": 2.5271,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.18276762402088773,
|
| 104 |
+
"grad_norm": 31.077625274658203,
|
| 105 |
+
"learning_rate": 4.6953872932985203e-05,
|
| 106 |
+
"loss": 2.6357,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.195822454308094,
|
| 111 |
+
"grad_norm": 24.155317306518555,
|
| 112 |
+
"learning_rate": 4.673629242819844e-05,
|
| 113 |
+
"loss": 2.5678,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.20887728459530025,
|
| 118 |
+
"grad_norm": 30.788408279418945,
|
| 119 |
+
"learning_rate": 4.651871192341166e-05,
|
| 120 |
+
"loss": 2.8182,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.22193211488250653,
|
| 125 |
+
"grad_norm": 25.434738159179688,
|
| 126 |
+
"learning_rate": 4.630113141862489e-05,
|
| 127 |
+
"loss": 3.303,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.2349869451697128,
|
| 132 |
+
"grad_norm": 37.103668212890625,
|
| 133 |
+
"learning_rate": 4.608355091383813e-05,
|
| 134 |
+
"loss": 2.5012,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.24804177545691905,
|
| 139 |
+
"grad_norm": 70.19502258300781,
|
| 140 |
+
"learning_rate": 4.586597040905135e-05,
|
| 141 |
+
"loss": 2.812,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.26109660574412535,
|
| 146 |
+
"grad_norm": 37.93436050415039,
|
| 147 |
+
"learning_rate": 4.564838990426458e-05,
|
| 148 |
+
"loss": 2.6683,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2741514360313316,
|
| 153 |
+
"grad_norm": 149.83016967773438,
|
| 154 |
+
"learning_rate": 4.543080939947781e-05,
|
| 155 |
+
"loss": 2.9776,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.28720626631853785,
|
| 160 |
+
"grad_norm": 31.74551010131836,
|
| 161 |
+
"learning_rate": 4.521322889469104e-05,
|
| 162 |
+
"loss": 2.9663,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.3002610966057441,
|
| 167 |
+
"grad_norm": 34.0869255065918,
|
| 168 |
+
"learning_rate": 4.4995648389904266e-05,
|
| 169 |
+
"loss": 2.712,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3133159268929504,
|
| 174 |
+
"grad_norm": 28.85022735595703,
|
| 175 |
+
"learning_rate": 4.47780678851175e-05,
|
| 176 |
+
"loss": 2.3012,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.3263707571801567,
|
| 181 |
+
"grad_norm": 36.18962860107422,
|
| 182 |
+
"learning_rate": 4.456048738033072e-05,
|
| 183 |
+
"loss": 2.4339,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.3394255874673629,
|
| 188 |
+
"grad_norm": 27.252077102661133,
|
| 189 |
+
"learning_rate": 4.4342906875543956e-05,
|
| 190 |
+
"loss": 2.851,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.3524804177545692,
|
| 195 |
+
"grad_norm": 38.39606857299805,
|
| 196 |
+
"learning_rate": 4.4125326370757184e-05,
|
| 197 |
+
"loss": 2.6867,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.36553524804177545,
|
| 202 |
+
"grad_norm": 25.8907527923584,
|
| 203 |
+
"learning_rate": 4.390774586597041e-05,
|
| 204 |
+
"loss": 2.4467,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.3785900783289817,
|
| 209 |
+
"grad_norm": 24.98986053466797,
|
| 210 |
+
"learning_rate": 4.369016536118364e-05,
|
| 211 |
+
"loss": 3.0278,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.391644908616188,
|
| 216 |
+
"grad_norm": 24.536916732788086,
|
| 217 |
+
"learning_rate": 4.347258485639687e-05,
|
| 218 |
+
"loss": 2.3857,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.4046997389033943,
|
| 223 |
+
"grad_norm": 22.012798309326172,
|
| 224 |
+
"learning_rate": 4.3255004351610094e-05,
|
| 225 |
+
"loss": 2.1414,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.4177545691906005,
|
| 230 |
+
"grad_norm": 25.466167449951172,
|
| 231 |
+
"learning_rate": 4.303742384682333e-05,
|
| 232 |
+
"loss": 2.5714,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.4308093994778068,
|
| 237 |
+
"grad_norm": 31.785062789916992,
|
| 238 |
+
"learning_rate": 4.281984334203655e-05,
|
| 239 |
+
"loss": 2.4608,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.44386422976501305,
|
| 244 |
+
"grad_norm": 36.67721176147461,
|
| 245 |
+
"learning_rate": 4.2602262837249784e-05,
|
| 246 |
+
"loss": 2.787,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.45691906005221933,
|
| 251 |
+
"grad_norm": 49.02054214477539,
|
| 252 |
+
"learning_rate": 4.238468233246302e-05,
|
| 253 |
+
"loss": 2.7207,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.4699738903394256,
|
| 258 |
+
"grad_norm": 26.876636505126953,
|
| 259 |
+
"learning_rate": 4.216710182767624e-05,
|
| 260 |
+
"loss": 2.6041,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.4830287206266319,
|
| 265 |
+
"grad_norm": 34.956336975097656,
|
| 266 |
+
"learning_rate": 4.1949521322889474e-05,
|
| 267 |
+
"loss": 3.1905,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.4960835509138381,
|
| 272 |
+
"grad_norm": 35.72273254394531,
|
| 273 |
+
"learning_rate": 4.17319408181027e-05,
|
| 274 |
+
"loss": 2.121,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.5091383812010444,
|
| 279 |
+
"grad_norm": 28.895980834960938,
|
| 280 |
+
"learning_rate": 4.151436031331593e-05,
|
| 281 |
+
"loss": 2.6956,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.5221932114882507,
|
| 286 |
+
"grad_norm": 28.925390243530273,
|
| 287 |
+
"learning_rate": 4.129677980852916e-05,
|
| 288 |
+
"loss": 3.0316,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.5352480417754569,
|
| 293 |
+
"grad_norm": 34.79185485839844,
|
| 294 |
+
"learning_rate": 4.107919930374239e-05,
|
| 295 |
+
"loss": 3.0389,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5483028720626631,
|
| 300 |
+
"grad_norm": 30.246923446655273,
|
| 301 |
+
"learning_rate": 4.086161879895561e-05,
|
| 302 |
+
"loss": 2.8086,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.5613577023498695,
|
| 307 |
+
"grad_norm": 32.78372573852539,
|
| 308 |
+
"learning_rate": 4.064403829416885e-05,
|
| 309 |
+
"loss": 2.814,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5744125326370757,
|
| 314 |
+
"grad_norm": 24.346147537231445,
|
| 315 |
+
"learning_rate": 4.0426457789382075e-05,
|
| 316 |
+
"loss": 2.8039,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.587467362924282,
|
| 321 |
+
"grad_norm": 31.166654586791992,
|
| 322 |
+
"learning_rate": 4.02088772845953e-05,
|
| 323 |
+
"loss": 2.9612,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6005221932114883,
|
| 328 |
+
"grad_norm": 23.0938777923584,
|
| 329 |
+
"learning_rate": 3.999129677980853e-05,
|
| 330 |
+
"loss": 2.5463,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6135770234986945,
|
| 335 |
+
"grad_norm": 26.590911865234375,
|
| 336 |
+
"learning_rate": 3.977371627502176e-05,
|
| 337 |
+
"loss": 2.3543,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6266318537859008,
|
| 342 |
+
"grad_norm": 29.803422927856445,
|
| 343 |
+
"learning_rate": 3.9556135770234985e-05,
|
| 344 |
+
"loss": 2.7445,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.639686684073107,
|
| 349 |
+
"grad_norm": 46.66853713989258,
|
| 350 |
+
"learning_rate": 3.933855526544822e-05,
|
| 351 |
+
"loss": 2.4905,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.6527415143603134,
|
| 356 |
+
"grad_norm": 39.04319381713867,
|
| 357 |
+
"learning_rate": 3.912097476066145e-05,
|
| 358 |
+
"loss": 1.8311,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.6657963446475196,
|
| 363 |
+
"grad_norm": 30.50276756286621,
|
| 364 |
+
"learning_rate": 3.8903394255874675e-05,
|
| 365 |
+
"loss": 2.8152,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.6788511749347258,
|
| 370 |
+
"grad_norm": 37.25984191894531,
|
| 371 |
+
"learning_rate": 3.868581375108791e-05,
|
| 372 |
+
"loss": 2.4759,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.6919060052219321,
|
| 377 |
+
"grad_norm": 25.89512062072754,
|
| 378 |
+
"learning_rate": 3.846823324630113e-05,
|
| 379 |
+
"loss": 2.486,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.7049608355091384,
|
| 384 |
+
"grad_norm": 37.318450927734375,
|
| 385 |
+
"learning_rate": 3.8250652741514365e-05,
|
| 386 |
+
"loss": 2.77,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.7180156657963447,
|
| 391 |
+
"grad_norm": 34.45144271850586,
|
| 392 |
+
"learning_rate": 3.803307223672759e-05,
|
| 393 |
+
"loss": 2.2646,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.7310704960835509,
|
| 398 |
+
"grad_norm": 50.494144439697266,
|
| 399 |
+
"learning_rate": 3.781549173194082e-05,
|
| 400 |
+
"loss": 2.7686,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.7441253263707572,
|
| 405 |
+
"grad_norm": 29.753643035888672,
|
| 406 |
+
"learning_rate": 3.759791122715405e-05,
|
| 407 |
+
"loss": 2.6239,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7571801566579635,
|
| 412 |
+
"grad_norm": 39.54145431518555,
|
| 413 |
+
"learning_rate": 3.738033072236728e-05,
|
| 414 |
+
"loss": 2.4995,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7702349869451697,
|
| 419 |
+
"grad_norm": 36.82713317871094,
|
| 420 |
+
"learning_rate": 3.71627502175805e-05,
|
| 421 |
+
"loss": 3.0274,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.783289817232376,
|
| 426 |
+
"grad_norm": 36.62627410888672,
|
| 427 |
+
"learning_rate": 3.694516971279374e-05,
|
| 428 |
+
"loss": 2.2364,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.7963446475195822,
|
| 433 |
+
"grad_norm": 18.279882431030273,
|
| 434 |
+
"learning_rate": 3.6727589208006965e-05,
|
| 435 |
+
"loss": 2.3391,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.8093994778067886,
|
| 440 |
+
"grad_norm": 23.61455535888672,
|
| 441 |
+
"learning_rate": 3.651000870322019e-05,
|
| 442 |
+
"loss": 2.5222,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.8224543080939948,
|
| 447 |
+
"grad_norm": 32.03522872924805,
|
| 448 |
+
"learning_rate": 3.629242819843342e-05,
|
| 449 |
+
"loss": 2.2871,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.835509138381201,
|
| 454 |
+
"grad_norm": 45.24649429321289,
|
| 455 |
+
"learning_rate": 3.607484769364665e-05,
|
| 456 |
+
"loss": 2.6863,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.8485639686684073,
|
| 461 |
+
"grad_norm": 24.39188575744629,
|
| 462 |
+
"learning_rate": 3.5857267188859876e-05,
|
| 463 |
+
"loss": 2.6426,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.8616187989556136,
|
| 468 |
+
"grad_norm": 21.67547607421875,
|
| 469 |
+
"learning_rate": 3.563968668407311e-05,
|
| 470 |
+
"loss": 2.1157,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.8746736292428199,
|
| 475 |
+
"grad_norm": 24.245168685913086,
|
| 476 |
+
"learning_rate": 3.542210617928634e-05,
|
| 477 |
+
"loss": 2.3781,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.8877284595300261,
|
| 482 |
+
"grad_norm": 27.57684326171875,
|
| 483 |
+
"learning_rate": 3.5204525674499566e-05,
|
| 484 |
+
"loss": 2.7673,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.9007832898172323,
|
| 489 |
+
"grad_norm": 28.42872428894043,
|
| 490 |
+
"learning_rate": 3.49869451697128e-05,
|
| 491 |
+
"loss": 2.4503,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.9138381201044387,
|
| 496 |
+
"grad_norm": 39.387813568115234,
|
| 497 |
+
"learning_rate": 3.476936466492602e-05,
|
| 498 |
+
"loss": 2.1765,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.9268929503916449,
|
| 503 |
+
"grad_norm": 20.197811126708984,
|
| 504 |
+
"learning_rate": 3.4551784160139256e-05,
|
| 505 |
+
"loss": 2.555,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.9399477806788512,
|
| 510 |
+
"grad_norm": 22.066137313842773,
|
| 511 |
+
"learning_rate": 3.4334203655352484e-05,
|
| 512 |
+
"loss": 2.4827,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.9530026109660574,
|
| 517 |
+
"grad_norm": 32.67851638793945,
|
| 518 |
+
"learning_rate": 3.411662315056571e-05,
|
| 519 |
+
"loss": 2.762,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9660574412532638,
|
| 524 |
+
"grad_norm": 28.471988677978516,
|
| 525 |
+
"learning_rate": 3.389904264577894e-05,
|
| 526 |
+
"loss": 2.5872,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.97911227154047,
|
| 531 |
+
"grad_norm": 22.934885025024414,
|
| 532 |
+
"learning_rate": 3.368146214099217e-05,
|
| 533 |
+
"loss": 2.8826,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.9921671018276762,
|
| 538 |
+
"grad_norm": 24.063716888427734,
|
| 539 |
+
"learning_rate": 3.3463881636205394e-05,
|
| 540 |
+
"loss": 2.8909,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.0,
|
| 545 |
+
"eval_loss": 2.3283705711364746,
|
| 546 |
+
"eval_runtime": 12.1219,
|
| 547 |
+
"eval_samples_per_second": 112.276,
|
| 548 |
+
"eval_steps_per_second": 14.107,
|
| 549 |
+
"step": 766
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.0052219321148825,
|
| 553 |
+
"grad_norm": 36.15023422241211,
|
| 554 |
+
"learning_rate": 3.324630113141863e-05,
|
| 555 |
+
"loss": 2.5282,
|
| 556 |
+
"step": 770
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.0182767624020888,
|
| 560 |
+
"grad_norm": 35.99642562866211,
|
| 561 |
+
"learning_rate": 3.3028720626631856e-05,
|
| 562 |
+
"loss": 2.176,
|
| 563 |
+
"step": 780
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.031331592689295,
|
| 567 |
+
"grad_norm": 30.217031478881836,
|
| 568 |
+
"learning_rate": 3.2811140121845084e-05,
|
| 569 |
+
"loss": 2.2727,
|
| 570 |
+
"step": 790
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.0443864229765012,
|
| 574 |
+
"grad_norm": 29.16168212890625,
|
| 575 |
+
"learning_rate": 3.259355961705831e-05,
|
| 576 |
+
"loss": 2.0302,
|
| 577 |
+
"step": 800
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.0574412532637076,
|
| 581 |
+
"grad_norm": 25.400541305541992,
|
| 582 |
+
"learning_rate": 3.237597911227154e-05,
|
| 583 |
+
"loss": 2.3489,
|
| 584 |
+
"step": 810
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.0704960835509139,
|
| 588 |
+
"grad_norm": 21.281591415405273,
|
| 589 |
+
"learning_rate": 3.215839860748477e-05,
|
| 590 |
+
"loss": 2.3976,
|
| 591 |
+
"step": 820
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.08355091383812,
|
| 595 |
+
"grad_norm": 23.941238403320312,
|
| 596 |
+
"learning_rate": 3.1940818102698e-05,
|
| 597 |
+
"loss": 2.6169,
|
| 598 |
+
"step": 830
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.0966057441253263,
|
| 602 |
+
"grad_norm": 26.626665115356445,
|
| 603 |
+
"learning_rate": 3.172323759791123e-05,
|
| 604 |
+
"loss": 2.6102,
|
| 605 |
+
"step": 840
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.1096605744125327,
|
| 609 |
+
"grad_norm": 28.539621353149414,
|
| 610 |
+
"learning_rate": 3.150565709312446e-05,
|
| 611 |
+
"loss": 2.019,
|
| 612 |
+
"step": 850
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.122715404699739,
|
| 616 |
+
"grad_norm": 36.77280044555664,
|
| 617 |
+
"learning_rate": 3.128807658833769e-05,
|
| 618 |
+
"loss": 2.2338,
|
| 619 |
+
"step": 860
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.1357702349869452,
|
| 623 |
+
"grad_norm": 38.7175407409668,
|
| 624 |
+
"learning_rate": 3.107049608355091e-05,
|
| 625 |
+
"loss": 2.7259,
|
| 626 |
+
"step": 870
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.1488250652741514,
|
| 630 |
+
"grad_norm": 32.9740104675293,
|
| 631 |
+
"learning_rate": 3.085291557876415e-05,
|
| 632 |
+
"loss": 2.294,
|
| 633 |
+
"step": 880
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.1618798955613576,
|
| 637 |
+
"grad_norm": 35.01115036010742,
|
| 638 |
+
"learning_rate": 3.0635335073977374e-05,
|
| 639 |
+
"loss": 2.5392,
|
| 640 |
+
"step": 890
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.174934725848564,
|
| 644 |
+
"grad_norm": 40.960968017578125,
|
| 645 |
+
"learning_rate": 3.0417754569190606e-05,
|
| 646 |
+
"loss": 1.8062,
|
| 647 |
+
"step": 900
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.1879895561357703,
|
| 651 |
+
"grad_norm": 24.627063751220703,
|
| 652 |
+
"learning_rate": 3.020017406440383e-05,
|
| 653 |
+
"loss": 2.746,
|
| 654 |
+
"step": 910
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.2010443864229765,
|
| 658 |
+
"grad_norm": 28.861692428588867,
|
| 659 |
+
"learning_rate": 2.998259355961706e-05,
|
| 660 |
+
"loss": 2.1434,
|
| 661 |
+
"step": 920
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.2140992167101827,
|
| 665 |
+
"grad_norm": 30.32466697692871,
|
| 666 |
+
"learning_rate": 2.976501305483029e-05,
|
| 667 |
+
"loss": 2.8282,
|
| 668 |
+
"step": 930
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.227154046997389,
|
| 672 |
+
"grad_norm": 45.79476547241211,
|
| 673 |
+
"learning_rate": 2.954743255004352e-05,
|
| 674 |
+
"loss": 2.1972,
|
| 675 |
+
"step": 940
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.2402088772845954,
|
| 679 |
+
"grad_norm": 34.27708435058594,
|
| 680 |
+
"learning_rate": 2.9329852045256744e-05,
|
| 681 |
+
"loss": 3.0359,
|
| 682 |
+
"step": 950
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.2532637075718016,
|
| 686 |
+
"grad_norm": 33.62773513793945,
|
| 687 |
+
"learning_rate": 2.9112271540469975e-05,
|
| 688 |
+
"loss": 2.4254,
|
| 689 |
+
"step": 960
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.2663185378590078,
|
| 693 |
+
"grad_norm": 26.693920135498047,
|
| 694 |
+
"learning_rate": 2.8894691035683203e-05,
|
| 695 |
+
"loss": 1.883,
|
| 696 |
+
"step": 970
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.279373368146214,
|
| 700 |
+
"grad_norm": 36.45111083984375,
|
| 701 |
+
"learning_rate": 2.8677110530896434e-05,
|
| 702 |
+
"loss": 2.232,
|
| 703 |
+
"step": 980
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.2924281984334205,
|
| 707 |
+
"grad_norm": 34.38032913208008,
|
| 708 |
+
"learning_rate": 2.845953002610966e-05,
|
| 709 |
+
"loss": 2.4289,
|
| 710 |
+
"step": 990
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.3054830287206267,
|
| 714 |
+
"grad_norm": 19.068925857543945,
|
| 715 |
+
"learning_rate": 2.8241949521322892e-05,
|
| 716 |
+
"loss": 2.473,
|
| 717 |
+
"step": 1000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.318537859007833,
|
| 721 |
+
"grad_norm": 30.37474250793457,
|
| 722 |
+
"learning_rate": 2.8024369016536117e-05,
|
| 723 |
+
"loss": 2.6123,
|
| 724 |
+
"step": 1010
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.3315926892950392,
|
| 728 |
+
"grad_norm": 29.176149368286133,
|
| 729 |
+
"learning_rate": 2.7806788511749348e-05,
|
| 730 |
+
"loss": 2.3158,
|
| 731 |
+
"step": 1020
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.3446475195822454,
|
| 735 |
+
"grad_norm": 27.721553802490234,
|
| 736 |
+
"learning_rate": 2.758920800696258e-05,
|
| 737 |
+
"loss": 2.3646,
|
| 738 |
+
"step": 1030
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.3577023498694518,
|
| 742 |
+
"grad_norm": 12.712676048278809,
|
| 743 |
+
"learning_rate": 2.7371627502175807e-05,
|
| 744 |
+
"loss": 2.1114,
|
| 745 |
+
"step": 1040
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.370757180156658,
|
| 749 |
+
"grad_norm": 23.869230270385742,
|
| 750 |
+
"learning_rate": 2.7154046997389038e-05,
|
| 751 |
+
"loss": 2.1335,
|
| 752 |
+
"step": 1050
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.3838120104438643,
|
| 756 |
+
"grad_norm": 23.66261100769043,
|
| 757 |
+
"learning_rate": 2.6936466492602262e-05,
|
| 758 |
+
"loss": 2.0774,
|
| 759 |
+
"step": 1060
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.3968668407310705,
|
| 763 |
+
"grad_norm": 44.03451156616211,
|
| 764 |
+
"learning_rate": 2.6718885987815496e-05,
|
| 765 |
+
"loss": 2.6329,
|
| 766 |
+
"step": 1070
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.4099216710182767,
|
| 770 |
+
"grad_norm": 20.849573135375977,
|
| 771 |
+
"learning_rate": 2.650130548302872e-05,
|
| 772 |
+
"loss": 2.8549,
|
| 773 |
+
"step": 1080
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.4229765013054831,
|
| 777 |
+
"grad_norm": 27.392250061035156,
|
| 778 |
+
"learning_rate": 2.6283724978241952e-05,
|
| 779 |
+
"loss": 2.4537,
|
| 780 |
+
"step": 1090
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.4360313315926894,
|
| 784 |
+
"grad_norm": 25.04733657836914,
|
| 785 |
+
"learning_rate": 2.606614447345518e-05,
|
| 786 |
+
"loss": 1.9425,
|
| 787 |
+
"step": 1100
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.4490861618798956,
|
| 791 |
+
"grad_norm": 32.247894287109375,
|
| 792 |
+
"learning_rate": 2.584856396866841e-05,
|
| 793 |
+
"loss": 2.0056,
|
| 794 |
+
"step": 1110
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.4621409921671018,
|
| 798 |
+
"grad_norm": 32.498191833496094,
|
| 799 |
+
"learning_rate": 2.5630983463881635e-05,
|
| 800 |
+
"loss": 2.6457,
|
| 801 |
+
"step": 1120
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.475195822454308,
|
| 805 |
+
"grad_norm": 26.728214263916016,
|
| 806 |
+
"learning_rate": 2.5413402959094866e-05,
|
| 807 |
+
"loss": 1.6954,
|
| 808 |
+
"step": 1130
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.4882506527415145,
|
| 812 |
+
"grad_norm": 63.32724380493164,
|
| 813 |
+
"learning_rate": 2.5195822454308094e-05,
|
| 814 |
+
"loss": 2.5008,
|
| 815 |
+
"step": 1140
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.5013054830287205,
|
| 819 |
+
"grad_norm": 24.65690803527832,
|
| 820 |
+
"learning_rate": 2.4978241949521325e-05,
|
| 821 |
+
"loss": 1.5901,
|
| 822 |
+
"step": 1150
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.514360313315927,
|
| 826 |
+
"grad_norm": 19.37055206298828,
|
| 827 |
+
"learning_rate": 2.4760661444734552e-05,
|
| 828 |
+
"loss": 1.9548,
|
| 829 |
+
"step": 1160
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.5274151436031331,
|
| 833 |
+
"grad_norm": 23.001806259155273,
|
| 834 |
+
"learning_rate": 2.4543080939947783e-05,
|
| 835 |
+
"loss": 2.382,
|
| 836 |
+
"step": 1170
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.5404699738903394,
|
| 840 |
+
"grad_norm": 26.997100830078125,
|
| 841 |
+
"learning_rate": 2.432550043516101e-05,
|
| 842 |
+
"loss": 2.2834,
|
| 843 |
+
"step": 1180
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.5535248041775458,
|
| 847 |
+
"grad_norm": 22.0489559173584,
|
| 848 |
+
"learning_rate": 2.410791993037424e-05,
|
| 849 |
+
"loss": 2.1768,
|
| 850 |
+
"step": 1190
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.566579634464752,
|
| 854 |
+
"grad_norm": 29.986967086791992,
|
| 855 |
+
"learning_rate": 2.389033942558747e-05,
|
| 856 |
+
"loss": 2.0854,
|
| 857 |
+
"step": 1200
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.5796344647519582,
|
| 861 |
+
"grad_norm": 34.0990104675293,
|
| 862 |
+
"learning_rate": 2.3672758920800698e-05,
|
| 863 |
+
"loss": 2.4429,
|
| 864 |
+
"step": 1210
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.5926892950391645,
|
| 868 |
+
"grad_norm": 25.31661033630371,
|
| 869 |
+
"learning_rate": 2.3455178416013925e-05,
|
| 870 |
+
"loss": 2.3145,
|
| 871 |
+
"step": 1220
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.6057441253263707,
|
| 875 |
+
"grad_norm": 30.300716400146484,
|
| 876 |
+
"learning_rate": 2.3237597911227156e-05,
|
| 877 |
+
"loss": 1.9357,
|
| 878 |
+
"step": 1230
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.6187989556135771,
|
| 882 |
+
"grad_norm": 25.158327102661133,
|
| 883 |
+
"learning_rate": 2.3020017406440384e-05,
|
| 884 |
+
"loss": 2.3195,
|
| 885 |
+
"step": 1240
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.6318537859007833,
|
| 889 |
+
"grad_norm": 33.35712432861328,
|
| 890 |
+
"learning_rate": 2.280243690165361e-05,
|
| 891 |
+
"loss": 2.4724,
|
| 892 |
+
"step": 1250
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.6449086161879896,
|
| 896 |
+
"grad_norm": 22.938852310180664,
|
| 897 |
+
"learning_rate": 2.258485639686684e-05,
|
| 898 |
+
"loss": 2.3876,
|
| 899 |
+
"step": 1260
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.6579634464751958,
|
| 903 |
+
"grad_norm": 26.44889259338379,
|
| 904 |
+
"learning_rate": 2.236727589208007e-05,
|
| 905 |
+
"loss": 2.1164,
|
| 906 |
+
"step": 1270
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.671018276762402,
|
| 910 |
+
"grad_norm": 28.251296997070312,
|
| 911 |
+
"learning_rate": 2.2149695387293298e-05,
|
| 912 |
+
"loss": 2.3045,
|
| 913 |
+
"step": 1280
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.6840731070496084,
|
| 917 |
+
"grad_norm": 28.00015640258789,
|
| 918 |
+
"learning_rate": 2.193211488250653e-05,
|
| 919 |
+
"loss": 1.9546,
|
| 920 |
+
"step": 1290
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.6971279373368147,
|
| 924 |
+
"grad_norm": 20.0263729095459,
|
| 925 |
+
"learning_rate": 2.171453437771976e-05,
|
| 926 |
+
"loss": 2.2491,
|
| 927 |
+
"step": 1300
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.7101827676240209,
|
| 931 |
+
"grad_norm": 23.335580825805664,
|
| 932 |
+
"learning_rate": 2.1496953872932988e-05,
|
| 933 |
+
"loss": 2.5543,
|
| 934 |
+
"step": 1310
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.723237597911227,
|
| 938 |
+
"grad_norm": 25.97711181640625,
|
| 939 |
+
"learning_rate": 2.1279373368146216e-05,
|
| 940 |
+
"loss": 2.2486,
|
| 941 |
+
"step": 1320
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.7362924281984333,
|
| 945 |
+
"grad_norm": 24.391855239868164,
|
| 946 |
+
"learning_rate": 2.1061792863359443e-05,
|
| 947 |
+
"loss": 2.3483,
|
| 948 |
+
"step": 1330
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.7493472584856398,
|
| 952 |
+
"grad_norm": 29.249792098999023,
|
| 953 |
+
"learning_rate": 2.0844212358572674e-05,
|
| 954 |
+
"loss": 2.3703,
|
| 955 |
+
"step": 1340
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.762402088772846,
|
| 959 |
+
"grad_norm": 20.213987350463867,
|
| 960 |
+
"learning_rate": 2.0626631853785902e-05,
|
| 961 |
+
"loss": 2.3178,
|
| 962 |
+
"step": 1350
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.7754569190600522,
|
| 966 |
+
"grad_norm": 33.050018310546875,
|
| 967 |
+
"learning_rate": 2.040905134899913e-05,
|
| 968 |
+
"loss": 2.2622,
|
| 969 |
+
"step": 1360
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.7885117493472587,
|
| 973 |
+
"grad_norm": 25.058115005493164,
|
| 974 |
+
"learning_rate": 2.019147084421236e-05,
|
| 975 |
+
"loss": 2.2577,
|
| 976 |
+
"step": 1370
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.8015665796344646,
|
| 980 |
+
"grad_norm": 34.79226303100586,
|
| 981 |
+
"learning_rate": 1.997389033942559e-05,
|
| 982 |
+
"loss": 2.1586,
|
| 983 |
+
"step": 1380
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.814621409921671,
|
| 987 |
+
"grad_norm": 38.56571578979492,
|
| 988 |
+
"learning_rate": 1.9756309834638816e-05,
|
| 989 |
+
"loss": 2.0911,
|
| 990 |
+
"step": 1390
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.8276762402088773,
|
| 994 |
+
"grad_norm": 27.778825759887695,
|
| 995 |
+
"learning_rate": 1.9538729329852047e-05,
|
| 996 |
+
"loss": 2.2379,
|
| 997 |
+
"step": 1400
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.8407310704960835,
|
| 1001 |
+
"grad_norm": 27.160274505615234,
|
| 1002 |
+
"learning_rate": 1.9321148825065275e-05,
|
| 1003 |
+
"loss": 2.0847,
|
| 1004 |
+
"step": 1410
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.85378590078329,
|
| 1008 |
+
"grad_norm": 26.11197853088379,
|
| 1009 |
+
"learning_rate": 1.9103568320278503e-05,
|
| 1010 |
+
"loss": 2.7021,
|
| 1011 |
+
"step": 1420
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.866840731070496,
|
| 1015 |
+
"grad_norm": 28.448244094848633,
|
| 1016 |
+
"learning_rate": 1.8885987815491734e-05,
|
| 1017 |
+
"loss": 2.1596,
|
| 1018 |
+
"step": 1430
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.8798955613577024,
|
| 1022 |
+
"grad_norm": 49.074729919433594,
|
| 1023 |
+
"learning_rate": 1.866840731070496e-05,
|
| 1024 |
+
"loss": 2.5753,
|
| 1025 |
+
"step": 1440
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.8929503916449086,
|
| 1029 |
+
"grad_norm": 21.96980094909668,
|
| 1030 |
+
"learning_rate": 1.845082680591819e-05,
|
| 1031 |
+
"loss": 2.5118,
|
| 1032 |
+
"step": 1450
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.9060052219321149,
|
| 1036 |
+
"grad_norm": 20.993181228637695,
|
| 1037 |
+
"learning_rate": 1.823324630113142e-05,
|
| 1038 |
+
"loss": 2.2332,
|
| 1039 |
+
"step": 1460
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.9190600522193213,
|
| 1043 |
+
"grad_norm": 20.049209594726562,
|
| 1044 |
+
"learning_rate": 1.801566579634465e-05,
|
| 1045 |
+
"loss": 2.1748,
|
| 1046 |
+
"step": 1470
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.9321148825065273,
|
| 1050 |
+
"grad_norm": 35.51521682739258,
|
| 1051 |
+
"learning_rate": 1.779808529155788e-05,
|
| 1052 |
+
"loss": 2.781,
|
| 1053 |
+
"step": 1480
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.9451697127937337,
|
| 1057 |
+
"grad_norm": 25.36643409729004,
|
| 1058 |
+
"learning_rate": 1.7580504786771106e-05,
|
| 1059 |
+
"loss": 2.7489,
|
| 1060 |
+
"step": 1490
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.95822454308094,
|
| 1064 |
+
"grad_norm": 38.09309387207031,
|
| 1065 |
+
"learning_rate": 1.7362924281984334e-05,
|
| 1066 |
+
"loss": 2.1326,
|
| 1067 |
+
"step": 1500
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.9712793733681462,
|
| 1071 |
+
"grad_norm": 30.636632919311523,
|
| 1072 |
+
"learning_rate": 1.7145343777197565e-05,
|
| 1073 |
+
"loss": 2.2151,
|
| 1074 |
+
"step": 1510
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.9843342036553526,
|
| 1078 |
+
"grad_norm": 27.038352966308594,
|
| 1079 |
+
"learning_rate": 1.6927763272410793e-05,
|
| 1080 |
+
"loss": 2.4523,
|
| 1081 |
+
"step": 1520
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.9973890339425586,
|
| 1085 |
+
"grad_norm": 19.101573944091797,
|
| 1086 |
+
"learning_rate": 1.671018276762402e-05,
|
| 1087 |
+
"loss": 2.518,
|
| 1088 |
+
"step": 1530
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 2.0,
|
| 1092 |
+
"eval_loss": NaN,
|
| 1093 |
+
"eval_runtime": 11.9494,
|
| 1094 |
+
"eval_samples_per_second": 113.897,
|
| 1095 |
+
"eval_steps_per_second": 14.31,
|
| 1096 |
+
"step": 1532
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 2.010443864229765,
|
| 1100 |
+
"grad_norm": 28.463035583496094,
|
| 1101 |
+
"learning_rate": 1.649260226283725e-05,
|
| 1102 |
+
"loss": 1.8945,
|
| 1103 |
+
"step": 1540
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 2.023498694516971,
|
| 1107 |
+
"grad_norm": 30.520097732543945,
|
| 1108 |
+
"learning_rate": 1.627502175805048e-05,
|
| 1109 |
+
"loss": 2.3685,
|
| 1110 |
+
"step": 1550
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 2.0365535248041775,
|
| 1114 |
+
"grad_norm": 19.876482009887695,
|
| 1115 |
+
"learning_rate": 1.6057441253263707e-05,
|
| 1116 |
+
"loss": 2.1494,
|
| 1117 |
+
"step": 1560
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 2.049608355091384,
|
| 1121 |
+
"grad_norm": 23.423219680786133,
|
| 1122 |
+
"learning_rate": 1.5839860748476938e-05,
|
| 1123 |
+
"loss": 1.9791,
|
| 1124 |
+
"step": 1570
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 2.06266318537859,
|
| 1128 |
+
"grad_norm": 20.257450103759766,
|
| 1129 |
+
"learning_rate": 1.5622280243690166e-05,
|
| 1130 |
+
"loss": 2.2526,
|
| 1131 |
+
"step": 1580
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 2.0757180156657964,
|
| 1135 |
+
"grad_norm": 47.68708038330078,
|
| 1136 |
+
"learning_rate": 1.5404699738903393e-05,
|
| 1137 |
+
"loss": 1.9962,
|
| 1138 |
+
"step": 1590
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 2.0887728459530024,
|
| 1142 |
+
"grad_norm": 27.561660766601562,
|
| 1143 |
+
"learning_rate": 1.5187119234116623e-05,
|
| 1144 |
+
"loss": 1.6106,
|
| 1145 |
+
"step": 1600
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 2.101827676240209,
|
| 1149 |
+
"grad_norm": 26.832944869995117,
|
| 1150 |
+
"learning_rate": 1.4969538729329852e-05,
|
| 1151 |
+
"loss": 1.5015,
|
| 1152 |
+
"step": 1610
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 2.1148825065274153,
|
| 1156 |
+
"grad_norm": 45.05983352661133,
|
| 1157 |
+
"learning_rate": 1.475195822454308e-05,
|
| 1158 |
+
"loss": 1.8897,
|
| 1159 |
+
"step": 1620
|
| 1160 |
+
},
|
| 1161 |
+
{
|
| 1162 |
+
"epoch": 2.1279373368146213,
|
| 1163 |
+
"grad_norm": 19.24533462524414,
|
| 1164 |
+
"learning_rate": 1.4534377719756313e-05,
|
| 1165 |
+
"loss": 2.4187,
|
| 1166 |
+
"step": 1630
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 2.1409921671018277,
|
| 1170 |
+
"grad_norm": 29.476770401000977,
|
| 1171 |
+
"learning_rate": 1.431679721496954e-05,
|
| 1172 |
+
"loss": 2.1733,
|
| 1173 |
+
"step": 1640
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 2.1540469973890337,
|
| 1177 |
+
"grad_norm": 26.505355834960938,
|
| 1178 |
+
"learning_rate": 1.409921671018277e-05,
|
| 1179 |
+
"loss": 1.8432,
|
| 1180 |
+
"step": 1650
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 2.16710182767624,
|
| 1184 |
+
"grad_norm": 28.1693058013916,
|
| 1185 |
+
"learning_rate": 1.3881636205395997e-05,
|
| 1186 |
+
"loss": 2.1419,
|
| 1187 |
+
"step": 1660
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 2.1801566579634466,
|
| 1191 |
+
"grad_norm": 37.704498291015625,
|
| 1192 |
+
"learning_rate": 1.3664055700609227e-05,
|
| 1193 |
+
"loss": 2.7488,
|
| 1194 |
+
"step": 1670
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 2.1932114882506526,
|
| 1198 |
+
"grad_norm": 22.38772964477539,
|
| 1199 |
+
"learning_rate": 1.3446475195822456e-05,
|
| 1200 |
+
"loss": 2.3531,
|
| 1201 |
+
"step": 1680
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 2.206266318537859,
|
| 1205 |
+
"grad_norm": 22.58838653564453,
|
| 1206 |
+
"learning_rate": 1.3228894691035684e-05,
|
| 1207 |
+
"loss": 1.8415,
|
| 1208 |
+
"step": 1690
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 2.2193211488250655,
|
| 1212 |
+
"grad_norm": 30.01149559020996,
|
| 1213 |
+
"learning_rate": 1.3011314186248913e-05,
|
| 1214 |
+
"loss": 2.0757,
|
| 1215 |
+
"step": 1700
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 2.2323759791122715,
|
| 1219 |
+
"grad_norm": 23.964759826660156,
|
| 1220 |
+
"learning_rate": 1.2793733681462141e-05,
|
| 1221 |
+
"loss": 2.4956,
|
| 1222 |
+
"step": 1710
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 2.245430809399478,
|
| 1226 |
+
"grad_norm": 33.133541107177734,
|
| 1227 |
+
"learning_rate": 1.257615317667537e-05,
|
| 1228 |
+
"loss": 1.971,
|
| 1229 |
+
"step": 1720
|
| 1230 |
+
},
|
| 1231 |
+
{
|
| 1232 |
+
"epoch": 2.258485639686684,
|
| 1233 |
+
"grad_norm": 27.34188461303711,
|
| 1234 |
+
"learning_rate": 1.23585726718886e-05,
|
| 1235 |
+
"loss": 2.2084,
|
| 1236 |
+
"step": 1730
|
| 1237 |
+
},
|
| 1238 |
+
{
|
| 1239 |
+
"epoch": 2.2715404699738904,
|
| 1240 |
+
"grad_norm": 25.62513542175293,
|
| 1241 |
+
"learning_rate": 1.2140992167101827e-05,
|
| 1242 |
+
"loss": 1.9445,
|
| 1243 |
+
"step": 1740
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 2.2845953002610964,
|
| 1247 |
+
"grad_norm": 33.618385314941406,
|
| 1248 |
+
"learning_rate": 1.1923411662315057e-05,
|
| 1249 |
+
"loss": 2.4505,
|
| 1250 |
+
"step": 1750
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 2.297650130548303,
|
| 1254 |
+
"grad_norm": 25.787757873535156,
|
| 1255 |
+
"learning_rate": 1.1705831157528286e-05,
|
| 1256 |
+
"loss": 2.748,
|
| 1257 |
+
"step": 1760
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 2.3107049608355092,
|
| 1261 |
+
"grad_norm": 24.533018112182617,
|
| 1262 |
+
"learning_rate": 1.1488250652741515e-05,
|
| 1263 |
+
"loss": 1.496,
|
| 1264 |
+
"step": 1770
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 2.3237597911227152,
|
| 1268 |
+
"grad_norm": 41.41205596923828,
|
| 1269 |
+
"learning_rate": 1.1270670147954745e-05,
|
| 1270 |
+
"loss": 2.5213,
|
| 1271 |
+
"step": 1780
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 2.3368146214099217,
|
| 1275 |
+
"grad_norm": 20.163238525390625,
|
| 1276 |
+
"learning_rate": 1.1053089643167972e-05,
|
| 1277 |
+
"loss": 1.8429,
|
| 1278 |
+
"step": 1790
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 2.349869451697128,
|
| 1282 |
+
"grad_norm": 36.58127975463867,
|
| 1283 |
+
"learning_rate": 1.0835509138381202e-05,
|
| 1284 |
+
"loss": 2.617,
|
| 1285 |
+
"step": 1800
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 2.362924281984334,
|
| 1289 |
+
"grad_norm": 29.932636260986328,
|
| 1290 |
+
"learning_rate": 1.061792863359443e-05,
|
| 1291 |
+
"loss": 2.2462,
|
| 1292 |
+
"step": 1810
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 2.3759791122715406,
|
| 1296 |
+
"grad_norm": 20.780025482177734,
|
| 1297 |
+
"learning_rate": 1.0400348128807659e-05,
|
| 1298 |
+
"loss": 1.7878,
|
| 1299 |
+
"step": 1820
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 2.3890339425587466,
|
| 1303 |
+
"grad_norm": 26.663557052612305,
|
| 1304 |
+
"learning_rate": 1.0182767624020888e-05,
|
| 1305 |
+
"loss": 2.2757,
|
| 1306 |
+
"step": 1830
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 2.402088772845953,
|
| 1310 |
+
"grad_norm": 34.194435119628906,
|
| 1311 |
+
"learning_rate": 9.965187119234116e-06,
|
| 1312 |
+
"loss": 1.8959,
|
| 1313 |
+
"step": 1840
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 2.4151436031331595,
|
| 1317 |
+
"grad_norm": 13.468405723571777,
|
| 1318 |
+
"learning_rate": 9.747606614447347e-06,
|
| 1319 |
+
"loss": 2.2176,
|
| 1320 |
+
"step": 1850
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 2.4281984334203655,
|
| 1324 |
+
"grad_norm": 36.56803512573242,
|
| 1325 |
+
"learning_rate": 9.530026109660575e-06,
|
| 1326 |
+
"loss": 2.2373,
|
| 1327 |
+
"step": 1860
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 2.441253263707572,
|
| 1331 |
+
"grad_norm": 25.580917358398438,
|
| 1332 |
+
"learning_rate": 9.312445604873804e-06,
|
| 1333 |
+
"loss": 2.004,
|
| 1334 |
+
"step": 1870
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 2.454308093994778,
|
| 1338 |
+
"grad_norm": 47.98051071166992,
|
| 1339 |
+
"learning_rate": 9.094865100087033e-06,
|
| 1340 |
+
"loss": 2.09,
|
| 1341 |
+
"step": 1880
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 2.4673629242819843,
|
| 1345 |
+
"grad_norm": 44.3414421081543,
|
| 1346 |
+
"learning_rate": 8.877284595300261e-06,
|
| 1347 |
+
"loss": 2.4412,
|
| 1348 |
+
"step": 1890
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 2.480417754569191,
|
| 1352 |
+
"grad_norm": 22.451644897460938,
|
| 1353 |
+
"learning_rate": 8.65970409051349e-06,
|
| 1354 |
+
"loss": 1.5646,
|
| 1355 |
+
"step": 1900
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 2.493472584856397,
|
| 1359 |
+
"grad_norm": 21.15165138244629,
|
| 1360 |
+
"learning_rate": 8.442123585726718e-06,
|
| 1361 |
+
"loss": 1.5899,
|
| 1362 |
+
"step": 1910
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 2.506527415143603,
|
| 1366 |
+
"grad_norm": 49.44068145751953,
|
| 1367 |
+
"learning_rate": 8.224543080939948e-06,
|
| 1368 |
+
"loss": 2.0553,
|
| 1369 |
+
"step": 1920
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 2.5195822454308097,
|
| 1373 |
+
"grad_norm": 30.540063858032227,
|
| 1374 |
+
"learning_rate": 8.006962576153177e-06,
|
| 1375 |
+
"loss": 1.7215,
|
| 1376 |
+
"step": 1930
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 2.5326370757180157,
|
| 1380 |
+
"grad_norm": 29.058853149414062,
|
| 1381 |
+
"learning_rate": 7.789382071366406e-06,
|
| 1382 |
+
"loss": 1.7025,
|
| 1383 |
+
"step": 1940
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 2.5456919060052217,
|
| 1387 |
+
"grad_norm": 36.38139343261719,
|
| 1388 |
+
"learning_rate": 7.571801566579635e-06,
|
| 1389 |
+
"loss": 1.9976,
|
| 1390 |
+
"step": 1950
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 2.558746736292428,
|
| 1394 |
+
"grad_norm": 21.742773056030273,
|
| 1395 |
+
"learning_rate": 7.354221061792864e-06,
|
| 1396 |
+
"loss": 1.8528,
|
| 1397 |
+
"step": 1960
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 2.5718015665796345,
|
| 1401 |
+
"grad_norm": 22.30496597290039,
|
| 1402 |
+
"learning_rate": 7.136640557006093e-06,
|
| 1403 |
+
"loss": 1.726,
|
| 1404 |
+
"step": 1970
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 2.584856396866841,
|
| 1408 |
+
"grad_norm": 28.57794761657715,
|
| 1409 |
+
"learning_rate": 6.919060052219321e-06,
|
| 1410 |
+
"loss": 1.9179,
|
| 1411 |
+
"step": 1980
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 2.597911227154047,
|
| 1415 |
+
"grad_norm": 35.680999755859375,
|
| 1416 |
+
"learning_rate": 6.70147954743255e-06,
|
| 1417 |
+
"loss": 1.9789,
|
| 1418 |
+
"step": 1990
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 2.6109660574412534,
|
| 1422 |
+
"grad_norm": 41.53910827636719,
|
| 1423 |
+
"learning_rate": 6.483899042645779e-06,
|
| 1424 |
+
"loss": 2.3311,
|
| 1425 |
+
"step": 2000
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 2.6240208877284594,
|
| 1429 |
+
"grad_norm": 29.586671829223633,
|
| 1430 |
+
"learning_rate": 6.266318537859008e-06,
|
| 1431 |
+
"loss": 1.698,
|
| 1432 |
+
"step": 2010
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 2.637075718015666,
|
| 1436 |
+
"grad_norm": 26.552576065063477,
|
| 1437 |
+
"learning_rate": 6.048738033072237e-06,
|
| 1438 |
+
"loss": 2.4366,
|
| 1439 |
+
"step": 2020
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 2.6501305483028723,
|
| 1443 |
+
"grad_norm": 23.361190795898438,
|
| 1444 |
+
"learning_rate": 5.831157528285466e-06,
|
| 1445 |
+
"loss": 1.8842,
|
| 1446 |
+
"step": 2030
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 2.6631853785900783,
|
| 1450 |
+
"grad_norm": 17.02872085571289,
|
| 1451 |
+
"learning_rate": 5.613577023498695e-06,
|
| 1452 |
+
"loss": 1.9653,
|
| 1453 |
+
"step": 2040
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 2.6762402088772848,
|
| 1457 |
+
"grad_norm": 36.75905227661133,
|
| 1458 |
+
"learning_rate": 5.3959965187119235e-06,
|
| 1459 |
+
"loss": 2.3589,
|
| 1460 |
+
"step": 2050
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 2.6892950391644908,
|
| 1464 |
+
"grad_norm": 25.204286575317383,
|
| 1465 |
+
"learning_rate": 5.178416013925153e-06,
|
| 1466 |
+
"loss": 2.0147,
|
| 1467 |
+
"step": 2060
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 2.702349869451697,
|
| 1471 |
+
"grad_norm": 29.083358764648438,
|
| 1472 |
+
"learning_rate": 4.9608355091383814e-06,
|
| 1473 |
+
"loss": 2.0933,
|
| 1474 |
+
"step": 2070
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 2.7154046997389036,
|
| 1478 |
+
"grad_norm": 31.50925064086914,
|
| 1479 |
+
"learning_rate": 4.743255004351611e-06,
|
| 1480 |
+
"loss": 1.9098,
|
| 1481 |
+
"step": 2080
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 2.7284595300261096,
|
| 1485 |
+
"grad_norm": 21.71343421936035,
|
| 1486 |
+
"learning_rate": 4.525674499564839e-06,
|
| 1487 |
+
"loss": 1.984,
|
| 1488 |
+
"step": 2090
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 2.741514360313316,
|
| 1492 |
+
"grad_norm": 23.28816032409668,
|
| 1493 |
+
"learning_rate": 4.308093994778068e-06,
|
| 1494 |
+
"loss": 2.287,
|
| 1495 |
+
"step": 2100
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 2.754569190600522,
|
| 1499 |
+
"grad_norm": 28.7834529876709,
|
| 1500 |
+
"learning_rate": 4.090513489991297e-06,
|
| 1501 |
+
"loss": 2.7365,
|
| 1502 |
+
"step": 2110
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 2.7676240208877285,
|
| 1506 |
+
"grad_norm": 39.411312103271484,
|
| 1507 |
+
"learning_rate": 3.872932985204526e-06,
|
| 1508 |
+
"loss": 1.7105,
|
| 1509 |
+
"step": 2120
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 2.780678851174935,
|
| 1513 |
+
"grad_norm": 30.489765167236328,
|
| 1514 |
+
"learning_rate": 3.655352480417755e-06,
|
| 1515 |
+
"loss": 2.5384,
|
| 1516 |
+
"step": 2130
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 2.793733681462141,
|
| 1520 |
+
"grad_norm": 25.620206832885742,
|
| 1521 |
+
"learning_rate": 3.4377719756309837e-06,
|
| 1522 |
+
"loss": 2.1768,
|
| 1523 |
+
"step": 2140
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 2.8067885117493474,
|
| 1527 |
+
"grad_norm": 31.87867546081543,
|
| 1528 |
+
"learning_rate": 3.2201914708442126e-06,
|
| 1529 |
+
"loss": 1.8912,
|
| 1530 |
+
"step": 2150
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 2.8198433420365534,
|
| 1534 |
+
"grad_norm": 28.86083984375,
|
| 1535 |
+
"learning_rate": 3.0026109660574416e-06,
|
| 1536 |
+
"loss": 2.0129,
|
| 1537 |
+
"step": 2160
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 2.83289817232376,
|
| 1541 |
+
"grad_norm": 21.942968368530273,
|
| 1542 |
+
"learning_rate": 2.78503046127067e-06,
|
| 1543 |
+
"loss": 2.1114,
|
| 1544 |
+
"step": 2170
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 2.8459530026109663,
|
| 1548 |
+
"grad_norm": 42.69978713989258,
|
| 1549 |
+
"learning_rate": 2.5674499564838995e-06,
|
| 1550 |
+
"loss": 2.0096,
|
| 1551 |
+
"step": 2180
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 2.8590078328981723,
|
| 1555 |
+
"grad_norm": 24.366918563842773,
|
| 1556 |
+
"learning_rate": 2.349869451697128e-06,
|
| 1557 |
+
"loss": 2.0789,
|
| 1558 |
+
"step": 2190
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 2.8720626631853787,
|
| 1562 |
+
"grad_norm": 21.407636642456055,
|
| 1563 |
+
"learning_rate": 2.132288946910357e-06,
|
| 1564 |
+
"loss": 2.0312,
|
| 1565 |
+
"step": 2200
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 2.8851174934725847,
|
| 1569 |
+
"grad_norm": 31.635135650634766,
|
| 1570 |
+
"learning_rate": 1.914708442123586e-06,
|
| 1571 |
+
"loss": 1.8093,
|
| 1572 |
+
"step": 2210
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 2.898172323759791,
|
| 1576 |
+
"grad_norm": 27.037294387817383,
|
| 1577 |
+
"learning_rate": 1.6971279373368149e-06,
|
| 1578 |
+
"loss": 1.7357,
|
| 1579 |
+
"step": 2220
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 2.9112271540469976,
|
| 1583 |
+
"grad_norm": 23.957460403442383,
|
| 1584 |
+
"learning_rate": 1.4795474325500436e-06,
|
| 1585 |
+
"loss": 2.1237,
|
| 1586 |
+
"step": 2230
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 2.9242819843342036,
|
| 1590 |
+
"grad_norm": 27.358850479125977,
|
| 1591 |
+
"learning_rate": 1.2619669277632725e-06,
|
| 1592 |
+
"loss": 2.2897,
|
| 1593 |
+
"step": 2240
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 2.93733681462141,
|
| 1597 |
+
"grad_norm": 52.11824417114258,
|
| 1598 |
+
"learning_rate": 1.0443864229765013e-06,
|
| 1599 |
+
"loss": 2.3002,
|
| 1600 |
+
"step": 2250
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 2.950391644908616,
|
| 1604 |
+
"grad_norm": 35.283416748046875,
|
| 1605 |
+
"learning_rate": 8.268059181897302e-07,
|
| 1606 |
+
"loss": 2.1628,
|
| 1607 |
+
"step": 2260
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 2.9634464751958225,
|
| 1611 |
+
"grad_norm": 30.233379364013672,
|
| 1612 |
+
"learning_rate": 6.092254134029592e-07,
|
| 1613 |
+
"loss": 2.16,
|
| 1614 |
+
"step": 2270
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 2.976501305483029,
|
| 1618 |
+
"grad_norm": 30.20307159423828,
|
| 1619 |
+
"learning_rate": 3.91644908616188e-07,
|
| 1620 |
+
"loss": 2.1221,
|
| 1621 |
+
"step": 2280
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 2.989556135770235,
|
| 1625 |
+
"grad_norm": 17.428722381591797,
|
| 1626 |
+
"learning_rate": 1.7406440382941688e-07,
|
| 1627 |
+
"loss": 2.0945,
|
| 1628 |
+
"step": 2290
|
| 1629 |
+
}
|
| 1630 |
+
],
|
| 1631 |
+
"logging_steps": 10,
|
| 1632 |
+
"max_steps": 2298,
|
| 1633 |
+
"num_input_tokens_seen": 0,
|
| 1634 |
+
"num_train_epochs": 3,
|
| 1635 |
+
"save_steps": 500,
|
| 1636 |
+
"stateful_callbacks": {
|
| 1637 |
+
"TrainerControl": {
|
| 1638 |
+
"args": {
|
| 1639 |
+
"should_epoch_stop": false,
|
| 1640 |
+
"should_evaluate": false,
|
| 1641 |
+
"should_log": false,
|
| 1642 |
+
"should_save": true,
|
| 1643 |
+
"should_training_stop": true
|
| 1644 |
+
},
|
| 1645 |
+
"attributes": {}
|
| 1646 |
+
}
|
| 1647 |
+
},
|
| 1648 |
+
"total_flos": 2416656676552704.0,
|
| 1649 |
+
"train_batch_size": 16,
|
| 1650 |
+
"trial_name": null,
|
| 1651 |
+
"trial_params": null
|
| 1652 |
+
}
|
checkpoint-2298/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f86adb9de168a824dd2efa36c83bb2323ddc512b2622a8e0f1f31147371530a
|
| 3 |
+
size 5304
|
config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "dbmdz/bert-base-turkish-uncased",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"BertForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"torch_dtype": "float32",
|
| 21 |
+
"transformers_version": "4.48.2",
|
| 22 |
+
"type_vocab_size": 2,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 32000
|
| 25 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"pad_token_id": 0,
|
| 4 |
+
"transformers_version": "4.48.2"
|
| 5 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f4e3329028953d7cfc643ab3b75ca2c123e19c0e0625c51591d26c864159317
|
| 3 |
+
size 442627224
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"4": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"max_len": 512,
|
| 51 |
+
"model_max_length": 512,
|
| 52 |
+
"never_split": null,
|
| 53 |
+
"pad_token": "[PAD]",
|
| 54 |
+
"sep_token": "[SEP]",
|
| 55 |
+
"strip_accents": null,
|
| 56 |
+
"tokenize_chinese_chars": true,
|
| 57 |
+
"tokenizer_class": "BertTokenizer",
|
| 58 |
+
"unk_token": "[UNK]"
|
| 59 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6f86adb9de168a824dd2efa36c83bb2323ddc512b2622a8e0f1f31147371530a
|
| 3 |
+
size 5304
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|