Commit
•
ad0d2af
1
Parent(s):
a12ac55
initial commit
Browse files- config.json +55 -0
- generation_config.json +13 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1727 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/pegasus-cnn_dailymail",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"activation_function": "relu",
|
| 5 |
+
"add_bias_logits": false,
|
| 6 |
+
"add_final_layer_norm": true,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"PegasusForConditionalGeneration"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.1,
|
| 11 |
+
"bos_token_id": 0,
|
| 12 |
+
"classif_dropout": 0.0,
|
| 13 |
+
"d_model": 1024,
|
| 14 |
+
"decoder_attention_heads": 16,
|
| 15 |
+
"decoder_ffn_dim": 4096,
|
| 16 |
+
"decoder_layerdrop": 0.0,
|
| 17 |
+
"decoder_layers": 16,
|
| 18 |
+
"decoder_start_token_id": 0,
|
| 19 |
+
"dropout": 0.1,
|
| 20 |
+
"encoder_attention_heads": 16,
|
| 21 |
+
"encoder_ffn_dim": 4096,
|
| 22 |
+
"encoder_layerdrop": 0.0,
|
| 23 |
+
"encoder_layers": 16,
|
| 24 |
+
"eos_token_id": 1,
|
| 25 |
+
"extra_pos_embeddings": 1,
|
| 26 |
+
"forced_eos_token_id": 1,
|
| 27 |
+
"id2label": {
|
| 28 |
+
"0": "LABEL_0",
|
| 29 |
+
"1": "LABEL_1",
|
| 30 |
+
"2": "LABEL_2"
|
| 31 |
+
},
|
| 32 |
+
"init_std": 0.02,
|
| 33 |
+
"is_encoder_decoder": true,
|
| 34 |
+
"label2id": {
|
| 35 |
+
"LABEL_0": 0,
|
| 36 |
+
"LABEL_1": 1,
|
| 37 |
+
"LABEL_2": 2
|
| 38 |
+
},
|
| 39 |
+
"length_penalty": 0.8,
|
| 40 |
+
"max_length": 128,
|
| 41 |
+
"max_position_embeddings": 1024,
|
| 42 |
+
"min_length": 32,
|
| 43 |
+
"model_type": "pegasus",
|
| 44 |
+
"normalize_before": true,
|
| 45 |
+
"normalize_embedding": false,
|
| 46 |
+
"num_beams": 8,
|
| 47 |
+
"num_hidden_layers": 16,
|
| 48 |
+
"pad_token_id": 0,
|
| 49 |
+
"scale_embedding": true,
|
| 50 |
+
"static_position_embeddings": true,
|
| 51 |
+
"torch_dtype": "float32",
|
| 52 |
+
"transformers_version": "4.42.3",
|
| 53 |
+
"use_cache": true,
|
| 54 |
+
"vocab_size": 96103
|
| 55 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"decoder_start_token_id": 0,
|
| 5 |
+
"eos_token_id": 1,
|
| 6 |
+
"forced_eos_token_id": 1,
|
| 7 |
+
"length_penalty": 0.8,
|
| 8 |
+
"max_length": 128,
|
| 9 |
+
"min_length": 32,
|
| 10 |
+
"num_beams": 8,
|
| 11 |
+
"pad_token_id": 0,
|
| 12 |
+
"transformers_version": "4.42.3"
|
| 13 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9b655a467d969328102682544bb1d0c6f739572aa7febdeb93979047e041e6e9
|
| 3 |
+
size 2283652852
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:88012c7353fd6a622a09dc4b2df192057370a2a64e64df2b5130ccc6cdb4b36e
|
| 3 |
+
size 4550170737
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c42a9c827be43eb80197c9c3bcfeecd93a972b16dd827e35c7abc10889d3173c
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6fad69a4849089764b606f01bdfe3196a50b3d18141c36f6e06543c90dec0722
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1727 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 24236,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.00412609341475491,
|
| 13 |
+
"grad_norm": 14.695940971374512,
|
| 14 |
+
"learning_rate": 1e-05,
|
| 15 |
+
"loss": 8.626,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.00825218682950982,
|
| 20 |
+
"grad_norm": 19.695737838745117,
|
| 21 |
+
"learning_rate": 2e-05,
|
| 22 |
+
"loss": 7.5714,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01237828024426473,
|
| 27 |
+
"grad_norm": 7.499746799468994,
|
| 28 |
+
"learning_rate": 3e-05,
|
| 29 |
+
"loss": 6.9595,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01650437365901964,
|
| 34 |
+
"grad_norm": 22.69700813293457,
|
| 35 |
+
"learning_rate": 4e-05,
|
| 36 |
+
"loss": 5.6622,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.02063046707377455,
|
| 41 |
+
"grad_norm": 16.763093948364258,
|
| 42 |
+
"learning_rate": 5e-05,
|
| 43 |
+
"loss": 1.9119,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02475656048852946,
|
| 48 |
+
"grad_norm": 3.74697208404541,
|
| 49 |
+
"learning_rate": 4.978934951129087e-05,
|
| 50 |
+
"loss": 1.6244,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02888265390328437,
|
| 55 |
+
"grad_norm": 6.804798126220703,
|
| 56 |
+
"learning_rate": 4.957869902258174e-05,
|
| 57 |
+
"loss": 1.4867,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03300874731803928,
|
| 62 |
+
"grad_norm": 4.382748603820801,
|
| 63 |
+
"learning_rate": 4.93680485338726e-05,
|
| 64 |
+
"loss": 1.4434,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03713484073279419,
|
| 69 |
+
"grad_norm": 2.020482063293457,
|
| 70 |
+
"learning_rate": 4.9157398045163464e-05,
|
| 71 |
+
"loss": 1.3701,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.0412609341475491,
|
| 76 |
+
"grad_norm": 1.9225651025772095,
|
| 77 |
+
"learning_rate": 4.894674755645433e-05,
|
| 78 |
+
"loss": 1.383,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04538702756230401,
|
| 83 |
+
"grad_norm": 1.8698792457580566,
|
| 84 |
+
"learning_rate": 4.87360970677452e-05,
|
| 85 |
+
"loss": 1.3512,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04951312097705892,
|
| 90 |
+
"grad_norm": 4.490991592407227,
|
| 91 |
+
"learning_rate": 4.852544657903607e-05,
|
| 92 |
+
"loss": 1.3039,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.05363921439181383,
|
| 97 |
+
"grad_norm": 2.697434186935425,
|
| 98 |
+
"learning_rate": 4.831479609032693e-05,
|
| 99 |
+
"loss": 1.3085,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.05776530780656874,
|
| 104 |
+
"grad_norm": 3.3568286895751953,
|
| 105 |
+
"learning_rate": 4.81041456016178e-05,
|
| 106 |
+
"loss": 1.2915,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.06189140122132365,
|
| 111 |
+
"grad_norm": 2.012889862060547,
|
| 112 |
+
"learning_rate": 4.789349511290866e-05,
|
| 113 |
+
"loss": 1.2778,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.06601749463607856,
|
| 118 |
+
"grad_norm": 4.024045467376709,
|
| 119 |
+
"learning_rate": 4.768284462419953e-05,
|
| 120 |
+
"loss": 1.2688,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.07014358805083347,
|
| 125 |
+
"grad_norm": 2.241870880126953,
|
| 126 |
+
"learning_rate": 4.7472194135490394e-05,
|
| 127 |
+
"loss": 1.2592,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.07426968146558838,
|
| 132 |
+
"grad_norm": 3.7178213596343994,
|
| 133 |
+
"learning_rate": 4.7261543646781266e-05,
|
| 134 |
+
"loss": 1.2112,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07839577488034329,
|
| 139 |
+
"grad_norm": 2.036505937576294,
|
| 140 |
+
"learning_rate": 4.705089315807213e-05,
|
| 141 |
+
"loss": 1.2126,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.0825218682950982,
|
| 146 |
+
"grad_norm": 1.7717580795288086,
|
| 147 |
+
"learning_rate": 4.6840242669362997e-05,
|
| 148 |
+
"loss": 1.191,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.08664796170985312,
|
| 153 |
+
"grad_norm": 1.8078298568725586,
|
| 154 |
+
"learning_rate": 4.662959218065386e-05,
|
| 155 |
+
"loss": 1.199,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.09077405512460802,
|
| 160 |
+
"grad_norm": 4.067634582519531,
|
| 161 |
+
"learning_rate": 4.641894169194473e-05,
|
| 162 |
+
"loss": 1.1415,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.09490014853936293,
|
| 167 |
+
"grad_norm": 1.9144686460494995,
|
| 168 |
+
"learning_rate": 4.620829120323559e-05,
|
| 169 |
+
"loss": 1.1486,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.09902624195411784,
|
| 174 |
+
"grad_norm": 1.4644508361816406,
|
| 175 |
+
"learning_rate": 4.599764071452646e-05,
|
| 176 |
+
"loss": 1.1665,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.10315233536887275,
|
| 181 |
+
"grad_norm": 1.6876461505889893,
|
| 182 |
+
"learning_rate": 4.578699022581733e-05,
|
| 183 |
+
"loss": 1.1197,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.10727842878362766,
|
| 188 |
+
"grad_norm": 1.6274546384811401,
|
| 189 |
+
"learning_rate": 4.5576339737108196e-05,
|
| 190 |
+
"loss": 1.1505,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.11140452219838257,
|
| 195 |
+
"grad_norm": 2.2767255306243896,
|
| 196 |
+
"learning_rate": 4.5365689248399054e-05,
|
| 197 |
+
"loss": 1.1376,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.11553061561313747,
|
| 202 |
+
"grad_norm": 2.0346932411193848,
|
| 203 |
+
"learning_rate": 4.5155038759689926e-05,
|
| 204 |
+
"loss": 1.0817,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.1196567090278924,
|
| 209 |
+
"grad_norm": 1.5901857614517212,
|
| 210 |
+
"learning_rate": 4.494438827098079e-05,
|
| 211 |
+
"loss": 1.0919,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.1237828024426473,
|
| 216 |
+
"grad_norm": 2.1194183826446533,
|
| 217 |
+
"learning_rate": 4.473373778227166e-05,
|
| 218 |
+
"loss": 1.0813,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1279088958574022,
|
| 223 |
+
"grad_norm": 4.601478576660156,
|
| 224 |
+
"learning_rate": 4.452308729356252e-05,
|
| 225 |
+
"loss": 1.1204,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.13203498927215712,
|
| 230 |
+
"grad_norm": 2.3545944690704346,
|
| 231 |
+
"learning_rate": 4.431243680485339e-05,
|
| 232 |
+
"loss": 1.0488,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.13616108268691204,
|
| 237 |
+
"grad_norm": 1.468526005744934,
|
| 238 |
+
"learning_rate": 4.410178631614425e-05,
|
| 239 |
+
"loss": 1.1108,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.14028717610166694,
|
| 244 |
+
"grad_norm": 1.5844910144805908,
|
| 245 |
+
"learning_rate": 4.389113582743512e-05,
|
| 246 |
+
"loss": 1.1114,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.14441326951642186,
|
| 251 |
+
"grad_norm": 4.228846073150635,
|
| 252 |
+
"learning_rate": 4.368048533872599e-05,
|
| 253 |
+
"loss": 1.0961,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.14853936293117675,
|
| 258 |
+
"grad_norm": 1.4088937044143677,
|
| 259 |
+
"learning_rate": 4.3469834850016856e-05,
|
| 260 |
+
"loss": 1.0758,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.15266545634593168,
|
| 265 |
+
"grad_norm": 1.6613353490829468,
|
| 266 |
+
"learning_rate": 4.325918436130772e-05,
|
| 267 |
+
"loss": 1.0916,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.15679154976068657,
|
| 272 |
+
"grad_norm": 1.6648322343826294,
|
| 273 |
+
"learning_rate": 4.304853387259859e-05,
|
| 274 |
+
"loss": 1.0784,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1609176431754415,
|
| 279 |
+
"grad_norm": 1.387666940689087,
|
| 280 |
+
"learning_rate": 4.283788338388945e-05,
|
| 281 |
+
"loss": 1.0894,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1650437365901964,
|
| 286 |
+
"grad_norm": 1.416618824005127,
|
| 287 |
+
"learning_rate": 4.262723289518032e-05,
|
| 288 |
+
"loss": 1.0283,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1691698300049513,
|
| 293 |
+
"grad_norm": 1.569306492805481,
|
| 294 |
+
"learning_rate": 4.241658240647118e-05,
|
| 295 |
+
"loss": 1.0513,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.17329592341970623,
|
| 300 |
+
"grad_norm": 1.4747782945632935,
|
| 301 |
+
"learning_rate": 4.2205931917762055e-05,
|
| 302 |
+
"loss": 1.0531,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.17742201683446113,
|
| 307 |
+
"grad_norm": 2.1925017833709717,
|
| 308 |
+
"learning_rate": 4.199528142905292e-05,
|
| 309 |
+
"loss": 1.058,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.18154811024921605,
|
| 314 |
+
"grad_norm": 1.9729565382003784,
|
| 315 |
+
"learning_rate": 4.1784630940343786e-05,
|
| 316 |
+
"loss": 1.0806,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.18567420366397094,
|
| 321 |
+
"grad_norm": 1.5546541213989258,
|
| 322 |
+
"learning_rate": 4.1573980451634644e-05,
|
| 323 |
+
"loss": 1.0746,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.18980029707872587,
|
| 328 |
+
"grad_norm": 1.8433148860931396,
|
| 329 |
+
"learning_rate": 4.1363329962925517e-05,
|
| 330 |
+
"loss": 1.0527,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.19392639049348076,
|
| 335 |
+
"grad_norm": 1.3673489093780518,
|
| 336 |
+
"learning_rate": 4.115267947421638e-05,
|
| 337 |
+
"loss": 1.0772,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.19805248390823568,
|
| 342 |
+
"grad_norm": 1.8290094137191772,
|
| 343 |
+
"learning_rate": 4.094202898550725e-05,
|
| 344 |
+
"loss": 1.0424,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.2021785773229906,
|
| 349 |
+
"grad_norm": 2.0811445713043213,
|
| 350 |
+
"learning_rate": 4.073137849679812e-05,
|
| 351 |
+
"loss": 1.0498,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.2063046707377455,
|
| 356 |
+
"grad_norm": 1.2849047183990479,
|
| 357 |
+
"learning_rate": 4.0520728008088985e-05,
|
| 358 |
+
"loss": 1.0341,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.21043076415250042,
|
| 363 |
+
"grad_norm": 3.288480281829834,
|
| 364 |
+
"learning_rate": 4.0310077519379843e-05,
|
| 365 |
+
"loss": 1.0273,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.21455685756725532,
|
| 370 |
+
"grad_norm": 1.2297766208648682,
|
| 371 |
+
"learning_rate": 4.009942703067071e-05,
|
| 372 |
+
"loss": 1.0768,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.21868295098201024,
|
| 377 |
+
"grad_norm": 1.4169477224349976,
|
| 378 |
+
"learning_rate": 3.988877654196158e-05,
|
| 379 |
+
"loss": 1.0334,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.22280904439676513,
|
| 384 |
+
"grad_norm": 1.314010739326477,
|
| 385 |
+
"learning_rate": 3.9678126053252446e-05,
|
| 386 |
+
"loss": 1.0208,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.22693513781152005,
|
| 391 |
+
"grad_norm": 1.5176063776016235,
|
| 392 |
+
"learning_rate": 3.946747556454331e-05,
|
| 393 |
+
"loss": 1.0561,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.23106123122627495,
|
| 398 |
+
"grad_norm": 1.418303370475769,
|
| 399 |
+
"learning_rate": 3.9256825075834184e-05,
|
| 400 |
+
"loss": 1.0223,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.23518732464102987,
|
| 405 |
+
"grad_norm": 1.8820278644561768,
|
| 406 |
+
"learning_rate": 3.904617458712504e-05,
|
| 407 |
+
"loss": 1.0333,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.2393134180557848,
|
| 412 |
+
"grad_norm": 1.7610660791397095,
|
| 413 |
+
"learning_rate": 3.883552409841591e-05,
|
| 414 |
+
"loss": 1.007,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.2434395114705397,
|
| 419 |
+
"grad_norm": 7.877830982208252,
|
| 420 |
+
"learning_rate": 3.862487360970677e-05,
|
| 421 |
+
"loss": 1.0138,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.2475656048852946,
|
| 426 |
+
"grad_norm": 1.4319870471954346,
|
| 427 |
+
"learning_rate": 3.8414223120997645e-05,
|
| 428 |
+
"loss": 1.0103,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.2516916983000495,
|
| 433 |
+
"grad_norm": 1.4879227876663208,
|
| 434 |
+
"learning_rate": 3.820357263228851e-05,
|
| 435 |
+
"loss": 1.0219,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.2558177917148044,
|
| 440 |
+
"grad_norm": 1.3280787467956543,
|
| 441 |
+
"learning_rate": 3.7992922143579376e-05,
|
| 442 |
+
"loss": 1.0157,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.25994388512955935,
|
| 447 |
+
"grad_norm": 2.4915549755096436,
|
| 448 |
+
"learning_rate": 3.778227165487024e-05,
|
| 449 |
+
"loss": 1.0422,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.26406997854431424,
|
| 454 |
+
"grad_norm": 1.3016897439956665,
|
| 455 |
+
"learning_rate": 3.757162116616111e-05,
|
| 456 |
+
"loss": 1.004,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.26819607195906914,
|
| 461 |
+
"grad_norm": 1.722939372062683,
|
| 462 |
+
"learning_rate": 3.736097067745197e-05,
|
| 463 |
+
"loss": 1.0196,
|
| 464 |
+
"step": 6500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.2723221653738241,
|
| 468 |
+
"grad_norm": 1.4764331579208374,
|
| 469 |
+
"learning_rate": 3.715032018874284e-05,
|
| 470 |
+
"loss": 0.9871,
|
| 471 |
+
"step": 6600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.276448258788579,
|
| 475 |
+
"grad_norm": 1.344777226448059,
|
| 476 |
+
"learning_rate": 3.693966970003371e-05,
|
| 477 |
+
"loss": 1.0249,
|
| 478 |
+
"step": 6700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2805743522033339,
|
| 482 |
+
"grad_norm": 1.1756465435028076,
|
| 483 |
+
"learning_rate": 3.6729019211324575e-05,
|
| 484 |
+
"loss": 1.0506,
|
| 485 |
+
"step": 6800
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.28470044561808877,
|
| 489 |
+
"grad_norm": 1.3845124244689941,
|
| 490 |
+
"learning_rate": 3.6518368722615434e-05,
|
| 491 |
+
"loss": 1.0041,
|
| 492 |
+
"step": 6900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.2888265390328437,
|
| 496 |
+
"grad_norm": 1.074078917503357,
|
| 497 |
+
"learning_rate": 3.6307718233906306e-05,
|
| 498 |
+
"loss": 0.9849,
|
| 499 |
+
"step": 7000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2929526324475986,
|
| 503 |
+
"grad_norm": 2.0719516277313232,
|
| 504 |
+
"learning_rate": 3.609706774519717e-05,
|
| 505 |
+
"loss": 1.0022,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2970787258623535,
|
| 510 |
+
"grad_norm": 1.1381429433822632,
|
| 511 |
+
"learning_rate": 3.5886417256488037e-05,
|
| 512 |
+
"loss": 0.9409,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.30120481927710846,
|
| 517 |
+
"grad_norm": 1.2426626682281494,
|
| 518 |
+
"learning_rate": 3.56757667677789e-05,
|
| 519 |
+
"loss": 1.0225,
|
| 520 |
+
"step": 7300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.30533091269186335,
|
| 524 |
+
"grad_norm": 2.445568561553955,
|
| 525 |
+
"learning_rate": 3.5465116279069774e-05,
|
| 526 |
+
"loss": 0.9725,
|
| 527 |
+
"step": 7400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.30945700610661825,
|
| 531 |
+
"grad_norm": 1.2126537561416626,
|
| 532 |
+
"learning_rate": 3.525446579036063e-05,
|
| 533 |
+
"loss": 1.0005,
|
| 534 |
+
"step": 7500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.31358309952137314,
|
| 538 |
+
"grad_norm": 2.634969472885132,
|
| 539 |
+
"learning_rate": 3.50438153016515e-05,
|
| 540 |
+
"loss": 1.0079,
|
| 541 |
+
"step": 7600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3177091929361281,
|
| 545 |
+
"grad_norm": 1.4859946966171265,
|
| 546 |
+
"learning_rate": 3.483316481294237e-05,
|
| 547 |
+
"loss": 1.0192,
|
| 548 |
+
"step": 7700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.321835286350883,
|
| 552 |
+
"grad_norm": 1.3265373706817627,
|
| 553 |
+
"learning_rate": 3.4622514324233236e-05,
|
| 554 |
+
"loss": 0.9836,
|
| 555 |
+
"step": 7800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.3259613797656379,
|
| 559 |
+
"grad_norm": 1.569514513015747,
|
| 560 |
+
"learning_rate": 3.44118638355241e-05,
|
| 561 |
+
"loss": 1.002,
|
| 562 |
+
"step": 7900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.3300874731803928,
|
| 566 |
+
"grad_norm": 1.718145728111267,
|
| 567 |
+
"learning_rate": 3.4201213346814966e-05,
|
| 568 |
+
"loss": 0.9599,
|
| 569 |
+
"step": 8000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.3342135665951477,
|
| 573 |
+
"grad_norm": 1.2960829734802246,
|
| 574 |
+
"learning_rate": 3.399056285810583e-05,
|
| 575 |
+
"loss": 1.0286,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.3383396600099026,
|
| 580 |
+
"grad_norm": 1.3030658960342407,
|
| 581 |
+
"learning_rate": 3.37799123693967e-05,
|
| 582 |
+
"loss": 0.9592,
|
| 583 |
+
"step": 8200
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3424657534246575,
|
| 587 |
+
"grad_norm": 1.6679294109344482,
|
| 588 |
+
"learning_rate": 3.356926188068756e-05,
|
| 589 |
+
"loss": 0.9823,
|
| 590 |
+
"step": 8300
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.34659184683941247,
|
| 594 |
+
"grad_norm": 1.079559326171875,
|
| 595 |
+
"learning_rate": 3.335861139197843e-05,
|
| 596 |
+
"loss": 0.9749,
|
| 597 |
+
"step": 8400
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.35071794025416736,
|
| 601 |
+
"grad_norm": 1.27901029586792,
|
| 602 |
+
"learning_rate": 3.31479609032693e-05,
|
| 603 |
+
"loss": 0.9801,
|
| 604 |
+
"step": 8500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.35484403366892225,
|
| 608 |
+
"grad_norm": 1.292656421661377,
|
| 609 |
+
"learning_rate": 3.2937310414560165e-05,
|
| 610 |
+
"loss": 0.9824,
|
| 611 |
+
"step": 8600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.35897012708367715,
|
| 615 |
+
"grad_norm": 1.2524762153625488,
|
| 616 |
+
"learning_rate": 3.272665992585103e-05,
|
| 617 |
+
"loss": 0.943,
|
| 618 |
+
"step": 8700
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.3630962204984321,
|
| 622 |
+
"grad_norm": 2.4386353492736816,
|
| 623 |
+
"learning_rate": 3.2516009437141896e-05,
|
| 624 |
+
"loss": 0.9738,
|
| 625 |
+
"step": 8800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.367222313913187,
|
| 629 |
+
"grad_norm": 1.2332638502120972,
|
| 630 |
+
"learning_rate": 3.230535894843276e-05,
|
| 631 |
+
"loss": 0.9599,
|
| 632 |
+
"step": 8900
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3713484073279419,
|
| 636 |
+
"grad_norm": 1.3955186605453491,
|
| 637 |
+
"learning_rate": 3.209470845972363e-05,
|
| 638 |
+
"loss": 1.0027,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.37547450074269684,
|
| 643 |
+
"grad_norm": 1.7736716270446777,
|
| 644 |
+
"learning_rate": 3.188405797101449e-05,
|
| 645 |
+
"loss": 0.9568,
|
| 646 |
+
"step": 9100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.37960059415745173,
|
| 650 |
+
"grad_norm": 1.1282614469528198,
|
| 651 |
+
"learning_rate": 3.1673407482305364e-05,
|
| 652 |
+
"loss": 0.9892,
|
| 653 |
+
"step": 9200
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.3837266875722066,
|
| 657 |
+
"grad_norm": 4.226625442504883,
|
| 658 |
+
"learning_rate": 3.146275699359622e-05,
|
| 659 |
+
"loss": 0.957,
|
| 660 |
+
"step": 9300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3878527809869615,
|
| 664 |
+
"grad_norm": 1.3062007427215576,
|
| 665 |
+
"learning_rate": 3.125210650488709e-05,
|
| 666 |
+
"loss": 0.9702,
|
| 667 |
+
"step": 9400
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.39197887440171647,
|
| 671 |
+
"grad_norm": 1.5109843015670776,
|
| 672 |
+
"learning_rate": 3.104145601617796e-05,
|
| 673 |
+
"loss": 0.9406,
|
| 674 |
+
"step": 9500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.39610496781647136,
|
| 678 |
+
"grad_norm": 1.2154899835586548,
|
| 679 |
+
"learning_rate": 3.0830805527468826e-05,
|
| 680 |
+
"loss": 0.9532,
|
| 681 |
+
"step": 9600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.40023106123122626,
|
| 685 |
+
"grad_norm": 1.239396095275879,
|
| 686 |
+
"learning_rate": 3.062015503875969e-05,
|
| 687 |
+
"loss": 0.9484,
|
| 688 |
+
"step": 9700
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.4043571546459812,
|
| 692 |
+
"grad_norm": 1.3215525150299072,
|
| 693 |
+
"learning_rate": 3.0409504550050553e-05,
|
| 694 |
+
"loss": 0.978,
|
| 695 |
+
"step": 9800
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.4084832480607361,
|
| 699 |
+
"grad_norm": 1.149057149887085,
|
| 700 |
+
"learning_rate": 3.0198854061341425e-05,
|
| 701 |
+
"loss": 0.968,
|
| 702 |
+
"step": 9900
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.412609341475491,
|
| 706 |
+
"grad_norm": 1.271074652671814,
|
| 707 |
+
"learning_rate": 2.998820357263229e-05,
|
| 708 |
+
"loss": 0.9794,
|
| 709 |
+
"step": 10000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.4167354348902459,
|
| 713 |
+
"grad_norm": 1.0992262363433838,
|
| 714 |
+
"learning_rate": 2.9777553083923153e-05,
|
| 715 |
+
"loss": 0.9594,
|
| 716 |
+
"step": 10100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.42086152830500084,
|
| 720 |
+
"grad_norm": 1.1205365657806396,
|
| 721 |
+
"learning_rate": 2.9566902595214025e-05,
|
| 722 |
+
"loss": 0.9439,
|
| 723 |
+
"step": 10200
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.42498762171975574,
|
| 727 |
+
"grad_norm": 1.144080638885498,
|
| 728 |
+
"learning_rate": 2.935625210650489e-05,
|
| 729 |
+
"loss": 0.9745,
|
| 730 |
+
"step": 10300
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.42911371513451063,
|
| 734 |
+
"grad_norm": 3.2051868438720703,
|
| 735 |
+
"learning_rate": 2.9145601617795752e-05,
|
| 736 |
+
"loss": 0.9594,
|
| 737 |
+
"step": 10400
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.4332398085492656,
|
| 741 |
+
"grad_norm": 1.2232369184494019,
|
| 742 |
+
"learning_rate": 2.8934951129086618e-05,
|
| 743 |
+
"loss": 0.9644,
|
| 744 |
+
"step": 10500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.4373659019640205,
|
| 748 |
+
"grad_norm": 1.3971831798553467,
|
| 749 |
+
"learning_rate": 2.872430064037749e-05,
|
| 750 |
+
"loss": 0.987,
|
| 751 |
+
"step": 10600
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.44149199537877537,
|
| 755 |
+
"grad_norm": 1.1187039613723755,
|
| 756 |
+
"learning_rate": 2.8513650151668352e-05,
|
| 757 |
+
"loss": 0.9657,
|
| 758 |
+
"step": 10700
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.44561808879353026,
|
| 762 |
+
"grad_norm": 1.1717453002929688,
|
| 763 |
+
"learning_rate": 2.8302999662959217e-05,
|
| 764 |
+
"loss": 0.9363,
|
| 765 |
+
"step": 10800
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.4497441822082852,
|
| 769 |
+
"grad_norm": 1.4479399919509888,
|
| 770 |
+
"learning_rate": 2.809234917425009e-05,
|
| 771 |
+
"loss": 0.9428,
|
| 772 |
+
"step": 10900
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.4538702756230401,
|
| 776 |
+
"grad_norm": 1.1537368297576904,
|
| 777 |
+
"learning_rate": 2.788169868554095e-05,
|
| 778 |
+
"loss": 0.9654,
|
| 779 |
+
"step": 11000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.457996369037795,
|
| 783 |
+
"grad_norm": 1.9704123735427856,
|
| 784 |
+
"learning_rate": 2.7671048196831817e-05,
|
| 785 |
+
"loss": 0.944,
|
| 786 |
+
"step": 11100
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.4621224624525499,
|
| 790 |
+
"grad_norm": 1.3609466552734375,
|
| 791 |
+
"learning_rate": 2.7460397708122682e-05,
|
| 792 |
+
"loss": 0.9353,
|
| 793 |
+
"step": 11200
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.46624855586730485,
|
| 797 |
+
"grad_norm": 1.3835324048995972,
|
| 798 |
+
"learning_rate": 2.724974721941355e-05,
|
| 799 |
+
"loss": 0.9238,
|
| 800 |
+
"step": 11300
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.47037464928205974,
|
| 804 |
+
"grad_norm": 2.1749815940856934,
|
| 805 |
+
"learning_rate": 2.7039096730704416e-05,
|
| 806 |
+
"loss": 0.9215,
|
| 807 |
+
"step": 11400
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.47450074269681464,
|
| 811 |
+
"grad_norm": 1.941735863685608,
|
| 812 |
+
"learning_rate": 2.682844624199528e-05,
|
| 813 |
+
"loss": 0.9607,
|
| 814 |
+
"step": 11500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.4786268361115696,
|
| 818 |
+
"grad_norm": 1.9667292833328247,
|
| 819 |
+
"learning_rate": 2.661779575328615e-05,
|
| 820 |
+
"loss": 0.9493,
|
| 821 |
+
"step": 11600
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.4827529295263245,
|
| 825 |
+
"grad_norm": 1.1912260055541992,
|
| 826 |
+
"learning_rate": 2.6407145264577016e-05,
|
| 827 |
+
"loss": 0.9364,
|
| 828 |
+
"step": 11700
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.4868790229410794,
|
| 832 |
+
"grad_norm": 1.2728015184402466,
|
| 833 |
+
"learning_rate": 2.619649477586788e-05,
|
| 834 |
+
"loss": 0.9135,
|
| 835 |
+
"step": 11800
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.49100511635583427,
|
| 839 |
+
"grad_norm": 1.326409935951233,
|
| 840 |
+
"learning_rate": 2.5985844287158746e-05,
|
| 841 |
+
"loss": 0.9665,
|
| 842 |
+
"step": 11900
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.4951312097705892,
|
| 846 |
+
"grad_norm": 1.4567406177520752,
|
| 847 |
+
"learning_rate": 2.5775193798449615e-05,
|
| 848 |
+
"loss": 0.9733,
|
| 849 |
+
"step": 12000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.4992573031853441,
|
| 853 |
+
"grad_norm": 1.3147661685943604,
|
| 854 |
+
"learning_rate": 2.556454330974048e-05,
|
| 855 |
+
"loss": 0.9204,
|
| 856 |
+
"step": 12100
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.503383396600099,
|
| 860 |
+
"grad_norm": 1.6704838275909424,
|
| 861 |
+
"learning_rate": 2.5353892821031346e-05,
|
| 862 |
+
"loss": 0.9418,
|
| 863 |
+
"step": 12200
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.5075094900148539,
|
| 867 |
+
"grad_norm": 1.2493371963500977,
|
| 868 |
+
"learning_rate": 2.5143242332322215e-05,
|
| 869 |
+
"loss": 0.9441,
|
| 870 |
+
"step": 12300
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.5116355834296088,
|
| 874 |
+
"grad_norm": 1.2380743026733398,
|
| 875 |
+
"learning_rate": 2.493259184361308e-05,
|
| 876 |
+
"loss": 0.9642,
|
| 877 |
+
"step": 12400
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.5157616768443638,
|
| 881 |
+
"grad_norm": 1.487196922302246,
|
| 882 |
+
"learning_rate": 2.4721941354903942e-05,
|
| 883 |
+
"loss": 0.986,
|
| 884 |
+
"step": 12500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.5198877702591187,
|
| 888 |
+
"grad_norm": 1.2720383405685425,
|
| 889 |
+
"learning_rate": 2.451129086619481e-05,
|
| 890 |
+
"loss": 0.9614,
|
| 891 |
+
"step": 12600
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.5240138636738736,
|
| 895 |
+
"grad_norm": 1.3985182046890259,
|
| 896 |
+
"learning_rate": 2.4300640377485676e-05,
|
| 897 |
+
"loss": 0.9327,
|
| 898 |
+
"step": 12700
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.5281399570886285,
|
| 902 |
+
"grad_norm": 1.2555489540100098,
|
| 903 |
+
"learning_rate": 2.408998988877654e-05,
|
| 904 |
+
"loss": 0.9331,
|
| 905 |
+
"step": 12800
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.5322660505033834,
|
| 909 |
+
"grad_norm": 1.083095908164978,
|
| 910 |
+
"learning_rate": 2.387933940006741e-05,
|
| 911 |
+
"loss": 0.9706,
|
| 912 |
+
"step": 12900
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.5363921439181383,
|
| 916 |
+
"grad_norm": 3.2246696949005127,
|
| 917 |
+
"learning_rate": 2.3668688911358276e-05,
|
| 918 |
+
"loss": 0.9267,
|
| 919 |
+
"step": 13000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.5405182373328932,
|
| 923 |
+
"grad_norm": 1.2211159467697144,
|
| 924 |
+
"learning_rate": 2.345803842264914e-05,
|
| 925 |
+
"loss": 0.9315,
|
| 926 |
+
"step": 13100
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.5446443307476482,
|
| 930 |
+
"grad_norm": 1.3726495504379272,
|
| 931 |
+
"learning_rate": 2.3247387933940006e-05,
|
| 932 |
+
"loss": 0.9432,
|
| 933 |
+
"step": 13200
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.5487704241624031,
|
| 937 |
+
"grad_norm": 1.0996991395950317,
|
| 938 |
+
"learning_rate": 2.3036737445230875e-05,
|
| 939 |
+
"loss": 0.9213,
|
| 940 |
+
"step": 13300
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.552896517577158,
|
| 944 |
+
"grad_norm": 1.016136884689331,
|
| 945 |
+
"learning_rate": 2.282608695652174e-05,
|
| 946 |
+
"loss": 0.9625,
|
| 947 |
+
"step": 13400
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.5570226109919129,
|
| 951 |
+
"grad_norm": 1.1178189516067505,
|
| 952 |
+
"learning_rate": 2.2615436467812606e-05,
|
| 953 |
+
"loss": 0.9419,
|
| 954 |
+
"step": 13500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.5611487044066678,
|
| 958 |
+
"grad_norm": 1.1706444025039673,
|
| 959 |
+
"learning_rate": 2.2404785979103475e-05,
|
| 960 |
+
"loss": 0.885,
|
| 961 |
+
"step": 13600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5652747978214226,
|
| 965 |
+
"grad_norm": 1.4330129623413086,
|
| 966 |
+
"learning_rate": 2.2194135490394337e-05,
|
| 967 |
+
"loss": 0.9024,
|
| 968 |
+
"step": 13700
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5694008912361775,
|
| 972 |
+
"grad_norm": 2.2776172161102295,
|
| 973 |
+
"learning_rate": 2.1983485001685205e-05,
|
| 974 |
+
"loss": 0.933,
|
| 975 |
+
"step": 13800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.5735269846509325,
|
| 979 |
+
"grad_norm": 1.3359657526016235,
|
| 980 |
+
"learning_rate": 2.177283451297607e-05,
|
| 981 |
+
"loss": 0.8791,
|
| 982 |
+
"step": 13900
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5776530780656874,
|
| 986 |
+
"grad_norm": 1.1592367887496948,
|
| 987 |
+
"learning_rate": 2.1562184024266936e-05,
|
| 988 |
+
"loss": 0.9336,
|
| 989 |
+
"step": 14000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5817791714804423,
|
| 993 |
+
"grad_norm": 1.052618145942688,
|
| 994 |
+
"learning_rate": 2.13515335355578e-05,
|
| 995 |
+
"loss": 0.9303,
|
| 996 |
+
"step": 14100
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5859052648951972,
|
| 1000 |
+
"grad_norm": 1.2330833673477173,
|
| 1001 |
+
"learning_rate": 2.114088304684867e-05,
|
| 1002 |
+
"loss": 0.9247,
|
| 1003 |
+
"step": 14200
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5900313583099521,
|
| 1007 |
+
"grad_norm": 1.7336995601654053,
|
| 1008 |
+
"learning_rate": 2.0930232558139536e-05,
|
| 1009 |
+
"loss": 0.9078,
|
| 1010 |
+
"step": 14300
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.594157451724707,
|
| 1014 |
+
"grad_norm": 1.1562308073043823,
|
| 1015 |
+
"learning_rate": 2.07195820694304e-05,
|
| 1016 |
+
"loss": 0.905,
|
| 1017 |
+
"step": 14400
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.5982835451394619,
|
| 1021 |
+
"grad_norm": 1.3212171792984009,
|
| 1022 |
+
"learning_rate": 2.050893158072127e-05,
|
| 1023 |
+
"loss": 0.9457,
|
| 1024 |
+
"step": 14500
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.6024096385542169,
|
| 1028 |
+
"grad_norm": 1.5021255016326904,
|
| 1029 |
+
"learning_rate": 2.0298281092012135e-05,
|
| 1030 |
+
"loss": 0.9213,
|
| 1031 |
+
"step": 14600
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.6065357319689718,
|
| 1035 |
+
"grad_norm": 1.1142035722732544,
|
| 1036 |
+
"learning_rate": 2.0087630603303e-05,
|
| 1037 |
+
"loss": 0.8988,
|
| 1038 |
+
"step": 14700
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.6106618253837267,
|
| 1042 |
+
"grad_norm": 1.0887188911437988,
|
| 1043 |
+
"learning_rate": 1.9876980114593866e-05,
|
| 1044 |
+
"loss": 0.9579,
|
| 1045 |
+
"step": 14800
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.6147879187984816,
|
| 1049 |
+
"grad_norm": 1.5622923374176025,
|
| 1050 |
+
"learning_rate": 1.966632962588473e-05,
|
| 1051 |
+
"loss": 0.9206,
|
| 1052 |
+
"step": 14900
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.6189140122132365,
|
| 1056 |
+
"grad_norm": 1.4978774785995483,
|
| 1057 |
+
"learning_rate": 1.94556791371756e-05,
|
| 1058 |
+
"loss": 0.9292,
|
| 1059 |
+
"step": 15000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.6230401056279914,
|
| 1063 |
+
"grad_norm": 1.1494709253311157,
|
| 1064 |
+
"learning_rate": 1.9245028648466465e-05,
|
| 1065 |
+
"loss": 0.929,
|
| 1066 |
+
"step": 15100
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.6271661990427463,
|
| 1070 |
+
"grad_norm": 3.5858824253082275,
|
| 1071 |
+
"learning_rate": 1.903437815975733e-05,
|
| 1072 |
+
"loss": 0.9181,
|
| 1073 |
+
"step": 15200
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.6312922924575013,
|
| 1077 |
+
"grad_norm": 0.927173376083374,
|
| 1078 |
+
"learning_rate": 1.8823727671048196e-05,
|
| 1079 |
+
"loss": 0.9365,
|
| 1080 |
+
"step": 15300
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.6354183858722562,
|
| 1084 |
+
"grad_norm": 0.9943380355834961,
|
| 1085 |
+
"learning_rate": 1.8613077182339065e-05,
|
| 1086 |
+
"loss": 0.8974,
|
| 1087 |
+
"step": 15400
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.6395444792870111,
|
| 1091 |
+
"grad_norm": 1.4820857048034668,
|
| 1092 |
+
"learning_rate": 1.840242669362993e-05,
|
| 1093 |
+
"loss": 0.9066,
|
| 1094 |
+
"step": 15500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.643670572701766,
|
| 1098 |
+
"grad_norm": 1.3542896509170532,
|
| 1099 |
+
"learning_rate": 1.8191776204920796e-05,
|
| 1100 |
+
"loss": 0.9048,
|
| 1101 |
+
"step": 15600
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.6477966661165209,
|
| 1105 |
+
"grad_norm": 2.233414888381958,
|
| 1106 |
+
"learning_rate": 1.7981125716211664e-05,
|
| 1107 |
+
"loss": 0.899,
|
| 1108 |
+
"step": 15700
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.6519227595312758,
|
| 1112 |
+
"grad_norm": 1.0770349502563477,
|
| 1113 |
+
"learning_rate": 1.777047522750253e-05,
|
| 1114 |
+
"loss": 0.9135,
|
| 1115 |
+
"step": 15800
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.6560488529460307,
|
| 1119 |
+
"grad_norm": 1.1688830852508545,
|
| 1120 |
+
"learning_rate": 1.7559824738793395e-05,
|
| 1121 |
+
"loss": 0.8838,
|
| 1122 |
+
"step": 15900
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.6601749463607856,
|
| 1126 |
+
"grad_norm": 1.096822738647461,
|
| 1127 |
+
"learning_rate": 1.734917425008426e-05,
|
| 1128 |
+
"loss": 0.9325,
|
| 1129 |
+
"step": 16000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.6643010397755406,
|
| 1133 |
+
"grad_norm": 1.4621776342391968,
|
| 1134 |
+
"learning_rate": 1.713852376137513e-05,
|
| 1135 |
+
"loss": 0.9299,
|
| 1136 |
+
"step": 16100
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.6684271331902955,
|
| 1140 |
+
"grad_norm": 1.2400994300842285,
|
| 1141 |
+
"learning_rate": 1.692787327266599e-05,
|
| 1142 |
+
"loss": 0.8986,
|
| 1143 |
+
"step": 16200
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.6725532266050503,
|
| 1147 |
+
"grad_norm": 1.3540397882461548,
|
| 1148 |
+
"learning_rate": 1.671722278395686e-05,
|
| 1149 |
+
"loss": 0.9084,
|
| 1150 |
+
"step": 16300
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.6766793200198052,
|
| 1154 |
+
"grad_norm": 1.2045152187347412,
|
| 1155 |
+
"learning_rate": 1.6506572295247725e-05,
|
| 1156 |
+
"loss": 0.8943,
|
| 1157 |
+
"step": 16400
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6808054134345601,
|
| 1161 |
+
"grad_norm": 1.1521943807601929,
|
| 1162 |
+
"learning_rate": 1.629592180653859e-05,
|
| 1163 |
+
"loss": 0.9089,
|
| 1164 |
+
"step": 16500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.684931506849315,
|
| 1168 |
+
"grad_norm": 4.699136257171631,
|
| 1169 |
+
"learning_rate": 1.608527131782946e-05,
|
| 1170 |
+
"loss": 0.9169,
|
| 1171 |
+
"step": 16600
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.6890576002640699,
|
| 1175 |
+
"grad_norm": 1.3759478330612183,
|
| 1176 |
+
"learning_rate": 1.5874620829120325e-05,
|
| 1177 |
+
"loss": 0.9154,
|
| 1178 |
+
"step": 16700
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6931836936788249,
|
| 1182 |
+
"grad_norm": 1.2098520994186401,
|
| 1183 |
+
"learning_rate": 1.566397034041119e-05,
|
| 1184 |
+
"loss": 0.9264,
|
| 1185 |
+
"step": 16800
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6973097870935798,
|
| 1189 |
+
"grad_norm": 1.6775233745574951,
|
| 1190 |
+
"learning_rate": 1.5453319851702056e-05,
|
| 1191 |
+
"loss": 0.9309,
|
| 1192 |
+
"step": 16900
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.7014358805083347,
|
| 1196 |
+
"grad_norm": 1.0574172735214233,
|
| 1197 |
+
"learning_rate": 1.5242669362992923e-05,
|
| 1198 |
+
"loss": 0.8893,
|
| 1199 |
+
"step": 17000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.7055619739230896,
|
| 1203 |
+
"grad_norm": 1.035610318183899,
|
| 1204 |
+
"learning_rate": 1.503201887428379e-05,
|
| 1205 |
+
"loss": 0.9243,
|
| 1206 |
+
"step": 17100
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.7096880673378445,
|
| 1210 |
+
"grad_norm": 1.6291944980621338,
|
| 1211 |
+
"learning_rate": 1.4821368385574655e-05,
|
| 1212 |
+
"loss": 0.9158,
|
| 1213 |
+
"step": 17200
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.7138141607525994,
|
| 1217 |
+
"grad_norm": 1.2090740203857422,
|
| 1218 |
+
"learning_rate": 1.4610717896865522e-05,
|
| 1219 |
+
"loss": 0.9026,
|
| 1220 |
+
"step": 17300
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.7179402541673543,
|
| 1224 |
+
"grad_norm": 1.2179425954818726,
|
| 1225 |
+
"learning_rate": 1.4400067408156388e-05,
|
| 1226 |
+
"loss": 0.8943,
|
| 1227 |
+
"step": 17400
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.7220663475821093,
|
| 1231 |
+
"grad_norm": 1.2382631301879883,
|
| 1232 |
+
"learning_rate": 1.4189416919447255e-05,
|
| 1233 |
+
"loss": 0.9021,
|
| 1234 |
+
"step": 17500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.7261924409968642,
|
| 1238 |
+
"grad_norm": 2.4923956394195557,
|
| 1239 |
+
"learning_rate": 1.3978766430738118e-05,
|
| 1240 |
+
"loss": 0.9348,
|
| 1241 |
+
"step": 17600
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.7303185344116191,
|
| 1245 |
+
"grad_norm": 2.419496774673462,
|
| 1246 |
+
"learning_rate": 1.3768115942028985e-05,
|
| 1247 |
+
"loss": 0.9039,
|
| 1248 |
+
"step": 17700
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.734444627826374,
|
| 1252 |
+
"grad_norm": 1.2352160215377808,
|
| 1253 |
+
"learning_rate": 1.3557465453319854e-05,
|
| 1254 |
+
"loss": 0.9266,
|
| 1255 |
+
"step": 17800
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.7385707212411289,
|
| 1259 |
+
"grad_norm": 1.0967360734939575,
|
| 1260 |
+
"learning_rate": 1.3346814964610718e-05,
|
| 1261 |
+
"loss": 0.9046,
|
| 1262 |
+
"step": 17900
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.7426968146558838,
|
| 1266 |
+
"grad_norm": 1.0056049823760986,
|
| 1267 |
+
"learning_rate": 1.3136164475901585e-05,
|
| 1268 |
+
"loss": 0.9321,
|
| 1269 |
+
"step": 18000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.7468229080706387,
|
| 1273 |
+
"grad_norm": 1.9823698997497559,
|
| 1274 |
+
"learning_rate": 1.292551398719245e-05,
|
| 1275 |
+
"loss": 0.9151,
|
| 1276 |
+
"step": 18100
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.7509490014853937,
|
| 1280 |
+
"grad_norm": 1.651145577430725,
|
| 1281 |
+
"learning_rate": 1.2714863498483317e-05,
|
| 1282 |
+
"loss": 0.904,
|
| 1283 |
+
"step": 18200
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.7550750949001486,
|
| 1287 |
+
"grad_norm": 0.9505665302276611,
|
| 1288 |
+
"learning_rate": 1.2504213009774183e-05,
|
| 1289 |
+
"loss": 0.878,
|
| 1290 |
+
"step": 18300
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.7592011883149035,
|
| 1294 |
+
"grad_norm": 1.558278203010559,
|
| 1295 |
+
"learning_rate": 1.229356252106505e-05,
|
| 1296 |
+
"loss": 0.942,
|
| 1297 |
+
"step": 18400
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.7633272817296584,
|
| 1301 |
+
"grad_norm": 1.2101174592971802,
|
| 1302 |
+
"learning_rate": 1.2082912032355915e-05,
|
| 1303 |
+
"loss": 0.9034,
|
| 1304 |
+
"step": 18500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.7674533751444133,
|
| 1308 |
+
"grad_norm": 1.2382097244262695,
|
| 1309 |
+
"learning_rate": 1.1872261543646782e-05,
|
| 1310 |
+
"loss": 0.9119,
|
| 1311 |
+
"step": 18600
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.7715794685591681,
|
| 1315 |
+
"grad_norm": 1.1424338817596436,
|
| 1316 |
+
"learning_rate": 1.1661611054937648e-05,
|
| 1317 |
+
"loss": 0.9282,
|
| 1318 |
+
"step": 18700
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.775705561973923,
|
| 1322 |
+
"grad_norm": 1.0747746229171753,
|
| 1323 |
+
"learning_rate": 1.1450960566228513e-05,
|
| 1324 |
+
"loss": 0.9144,
|
| 1325 |
+
"step": 18800
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.779831655388678,
|
| 1329 |
+
"grad_norm": 1.4378238916397095,
|
| 1330 |
+
"learning_rate": 1.1240310077519382e-05,
|
| 1331 |
+
"loss": 0.9292,
|
| 1332 |
+
"step": 18900
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.7839577488034329,
|
| 1336 |
+
"grad_norm": 1.5118451118469238,
|
| 1337 |
+
"learning_rate": 1.1029659588810247e-05,
|
| 1338 |
+
"loss": 0.8532,
|
| 1339 |
+
"step": 19000
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.7880838422181878,
|
| 1343 |
+
"grad_norm": 1.135190725326538,
|
| 1344 |
+
"learning_rate": 1.0819009100101113e-05,
|
| 1345 |
+
"loss": 0.9132,
|
| 1346 |
+
"step": 19100
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.7922099356329427,
|
| 1350 |
+
"grad_norm": 1.3497545719146729,
|
| 1351 |
+
"learning_rate": 1.060835861139198e-05,
|
| 1352 |
+
"loss": 0.9214,
|
| 1353 |
+
"step": 19200
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.7963360290476976,
|
| 1357 |
+
"grad_norm": 1.3251924514770508,
|
| 1358 |
+
"learning_rate": 1.0397708122682845e-05,
|
| 1359 |
+
"loss": 0.8942,
|
| 1360 |
+
"step": 19300
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.8004621224624525,
|
| 1364 |
+
"grad_norm": 2.453803539276123,
|
| 1365 |
+
"learning_rate": 1.018705763397371e-05,
|
| 1366 |
+
"loss": 0.8858,
|
| 1367 |
+
"step": 19400
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.8045882158772074,
|
| 1371 |
+
"grad_norm": 1.1651134490966797,
|
| 1372 |
+
"learning_rate": 9.976407145264577e-06,
|
| 1373 |
+
"loss": 0.9012,
|
| 1374 |
+
"step": 19500
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.8087143092919624,
|
| 1378 |
+
"grad_norm": 2.257159471511841,
|
| 1379 |
+
"learning_rate": 9.765756656555444e-06,
|
| 1380 |
+
"loss": 0.9167,
|
| 1381 |
+
"step": 19600
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.8128404027067173,
|
| 1385 |
+
"grad_norm": 0.9240596294403076,
|
| 1386 |
+
"learning_rate": 9.55510616784631e-06,
|
| 1387 |
+
"loss": 0.8764,
|
| 1388 |
+
"step": 19700
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.8169664961214722,
|
| 1392 |
+
"grad_norm": 1.2550618648529053,
|
| 1393 |
+
"learning_rate": 9.344455679137177e-06,
|
| 1394 |
+
"loss": 0.8998,
|
| 1395 |
+
"step": 19800
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.8210925895362271,
|
| 1399 |
+
"grad_norm": 1.2276984453201294,
|
| 1400 |
+
"learning_rate": 9.133805190428042e-06,
|
| 1401 |
+
"loss": 0.909,
|
| 1402 |
+
"step": 19900
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.825218682950982,
|
| 1406 |
+
"grad_norm": 1.0953816175460815,
|
| 1407 |
+
"learning_rate": 8.923154701718908e-06,
|
| 1408 |
+
"loss": 0.8931,
|
| 1409 |
+
"step": 20000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.8293447763657369,
|
| 1413 |
+
"grad_norm": 1.469269037246704,
|
| 1414 |
+
"learning_rate": 8.712504213009775e-06,
|
| 1415 |
+
"loss": 0.8789,
|
| 1416 |
+
"step": 20100
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.8334708697804918,
|
| 1420 |
+
"grad_norm": 1.242390751838684,
|
| 1421 |
+
"learning_rate": 8.50185372430064e-06,
|
| 1422 |
+
"loss": 0.9126,
|
| 1423 |
+
"step": 20200
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.8375969631952468,
|
| 1427 |
+
"grad_norm": 1.0811703205108643,
|
| 1428 |
+
"learning_rate": 8.291203235591507e-06,
|
| 1429 |
+
"loss": 0.913,
|
| 1430 |
+
"step": 20300
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.8417230566100017,
|
| 1434 |
+
"grad_norm": 1.0523350238800049,
|
| 1435 |
+
"learning_rate": 8.080552746882374e-06,
|
| 1436 |
+
"loss": 0.9118,
|
| 1437 |
+
"step": 20400
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.8458491500247566,
|
| 1441 |
+
"grad_norm": 1.4592727422714233,
|
| 1442 |
+
"learning_rate": 7.86990225817324e-06,
|
| 1443 |
+
"loss": 0.9099,
|
| 1444 |
+
"step": 20500
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.8499752434395115,
|
| 1448 |
+
"grad_norm": 1.0648339986801147,
|
| 1449 |
+
"learning_rate": 7.659251769464105e-06,
|
| 1450 |
+
"loss": 0.9198,
|
| 1451 |
+
"step": 20600
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.8541013368542664,
|
| 1455 |
+
"grad_norm": 1.3053339719772339,
|
| 1456 |
+
"learning_rate": 7.448601280754971e-06,
|
| 1457 |
+
"loss": 0.9156,
|
| 1458 |
+
"step": 20700
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.8582274302690213,
|
| 1462 |
+
"grad_norm": 1.0929012298583984,
|
| 1463 |
+
"learning_rate": 7.237950792045837e-06,
|
| 1464 |
+
"loss": 0.9103,
|
| 1465 |
+
"step": 20800
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.8623535236837762,
|
| 1469 |
+
"grad_norm": 1.234263300895691,
|
| 1470 |
+
"learning_rate": 7.027300303336704e-06,
|
| 1471 |
+
"loss": 0.9016,
|
| 1472 |
+
"step": 20900
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.8664796170985312,
|
| 1476 |
+
"grad_norm": 1.3241745233535767,
|
| 1477 |
+
"learning_rate": 6.816649814627571e-06,
|
| 1478 |
+
"loss": 0.9075,
|
| 1479 |
+
"step": 21000
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.8706057105132861,
|
| 1483 |
+
"grad_norm": 1.2847357988357544,
|
| 1484 |
+
"learning_rate": 6.605999325918437e-06,
|
| 1485 |
+
"loss": 0.9462,
|
| 1486 |
+
"step": 21100
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.874731803928041,
|
| 1490 |
+
"grad_norm": 1.1206868886947632,
|
| 1491 |
+
"learning_rate": 6.395348837209303e-06,
|
| 1492 |
+
"loss": 0.9236,
|
| 1493 |
+
"step": 21200
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.8788578973427958,
|
| 1497 |
+
"grad_norm": 1.1748895645141602,
|
| 1498 |
+
"learning_rate": 6.1846983485001685e-06,
|
| 1499 |
+
"loss": 0.8521,
|
| 1500 |
+
"step": 21300
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.8829839907575507,
|
| 1504 |
+
"grad_norm": 1.571519136428833,
|
| 1505 |
+
"learning_rate": 5.974047859791035e-06,
|
| 1506 |
+
"loss": 0.8811,
|
| 1507 |
+
"step": 21400
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.8871100841723056,
|
| 1511 |
+
"grad_norm": 1.051316738128662,
|
| 1512 |
+
"learning_rate": 5.763397371081901e-06,
|
| 1513 |
+
"loss": 0.9167,
|
| 1514 |
+
"step": 21500
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.8912361775870605,
|
| 1518 |
+
"grad_norm": 2.111393690109253,
|
| 1519 |
+
"learning_rate": 5.552746882372767e-06,
|
| 1520 |
+
"loss": 0.9333,
|
| 1521 |
+
"step": 21600
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.8953622710018155,
|
| 1525 |
+
"grad_norm": 1.3969411849975586,
|
| 1526 |
+
"learning_rate": 5.342096393663633e-06,
|
| 1527 |
+
"loss": 0.9434,
|
| 1528 |
+
"step": 21700
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.8994883644165704,
|
| 1532 |
+
"grad_norm": 1.7783890962600708,
|
| 1533 |
+
"learning_rate": 5.1314459049545e-06,
|
| 1534 |
+
"loss": 0.8947,
|
| 1535 |
+
"step": 21800
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.9036144578313253,
|
| 1539 |
+
"grad_norm": 1.359174132347107,
|
| 1540 |
+
"learning_rate": 4.920795416245366e-06,
|
| 1541 |
+
"loss": 0.8815,
|
| 1542 |
+
"step": 21900
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.9077405512460802,
|
| 1546 |
+
"grad_norm": 1.257117748260498,
|
| 1547 |
+
"learning_rate": 4.710144927536232e-06,
|
| 1548 |
+
"loss": 0.8986,
|
| 1549 |
+
"step": 22000
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.9118666446608351,
|
| 1553 |
+
"grad_norm": 0.9748762845993042,
|
| 1554 |
+
"learning_rate": 4.499494438827098e-06,
|
| 1555 |
+
"loss": 0.887,
|
| 1556 |
+
"step": 22100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.91599273807559,
|
| 1560 |
+
"grad_norm": 1.5360727310180664,
|
| 1561 |
+
"learning_rate": 4.2888439501179645e-06,
|
| 1562 |
+
"loss": 0.9051,
|
| 1563 |
+
"step": 22200
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.9201188314903449,
|
| 1567 |
+
"grad_norm": 1.0747774839401245,
|
| 1568 |
+
"learning_rate": 4.078193461408831e-06,
|
| 1569 |
+
"loss": 0.9498,
|
| 1570 |
+
"step": 22300
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.9242449249050998,
|
| 1574 |
+
"grad_norm": 1.197403073310852,
|
| 1575 |
+
"learning_rate": 3.867542972699697e-06,
|
| 1576 |
+
"loss": 0.925,
|
| 1577 |
+
"step": 22400
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.9283710183198548,
|
| 1581 |
+
"grad_norm": 1.580825924873352,
|
| 1582 |
+
"learning_rate": 3.6568924839905627e-06,
|
| 1583 |
+
"loss": 0.9019,
|
| 1584 |
+
"step": 22500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.9324971117346097,
|
| 1588 |
+
"grad_norm": 1.2338446378707886,
|
| 1589 |
+
"learning_rate": 3.4462419952814294e-06,
|
| 1590 |
+
"loss": 0.892,
|
| 1591 |
+
"step": 22600
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.9366232051493646,
|
| 1595 |
+
"grad_norm": 2.6846702098846436,
|
| 1596 |
+
"learning_rate": 3.2355915065722956e-06,
|
| 1597 |
+
"loss": 0.9121,
|
| 1598 |
+
"step": 22700
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.9407492985641195,
|
| 1602 |
+
"grad_norm": 1.0765039920806885,
|
| 1603 |
+
"learning_rate": 3.0249410178631614e-06,
|
| 1604 |
+
"loss": 0.9101,
|
| 1605 |
+
"step": 22800
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.9448753919788744,
|
| 1609 |
+
"grad_norm": 1.2273006439208984,
|
| 1610 |
+
"learning_rate": 2.814290529154028e-06,
|
| 1611 |
+
"loss": 0.8916,
|
| 1612 |
+
"step": 22900
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.9490014853936293,
|
| 1616 |
+
"grad_norm": 1.295823574066162,
|
| 1617 |
+
"learning_rate": 2.603640040444894e-06,
|
| 1618 |
+
"loss": 0.8958,
|
| 1619 |
+
"step": 23000
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.9531275788083842,
|
| 1623 |
+
"grad_norm": 1.5502872467041016,
|
| 1624 |
+
"learning_rate": 2.39298955173576e-06,
|
| 1625 |
+
"loss": 0.9069,
|
| 1626 |
+
"step": 23100
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.9572536722231392,
|
| 1630 |
+
"grad_norm": 2.524392604827881,
|
| 1631 |
+
"learning_rate": 2.1823390630266263e-06,
|
| 1632 |
+
"loss": 0.9301,
|
| 1633 |
+
"step": 23200
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.9613797656378941,
|
| 1637 |
+
"grad_norm": 1.5065919160842896,
|
| 1638 |
+
"learning_rate": 1.9716885743174925e-06,
|
| 1639 |
+
"loss": 0.8903,
|
| 1640 |
+
"step": 23300
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.965505859052649,
|
| 1644 |
+
"grad_norm": 1.1356451511383057,
|
| 1645 |
+
"learning_rate": 1.761038085608359e-06,
|
| 1646 |
+
"loss": 0.881,
|
| 1647 |
+
"step": 23400
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.9696319524674039,
|
| 1651 |
+
"grad_norm": 6.048961162567139,
|
| 1652 |
+
"learning_rate": 1.550387596899225e-06,
|
| 1653 |
+
"loss": 0.9164,
|
| 1654 |
+
"step": 23500
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.9737580458821588,
|
| 1658 |
+
"grad_norm": 1.1151750087738037,
|
| 1659 |
+
"learning_rate": 1.339737108190091e-06,
|
| 1660 |
+
"loss": 0.8703,
|
| 1661 |
+
"step": 23600
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.9778841392969136,
|
| 1665 |
+
"grad_norm": 1.0021706819534302,
|
| 1666 |
+
"learning_rate": 1.1290866194809571e-06,
|
| 1667 |
+
"loss": 0.8978,
|
| 1668 |
+
"step": 23700
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.9820102327116685,
|
| 1672 |
+
"grad_norm": 1.1401609182357788,
|
| 1673 |
+
"learning_rate": 9.184361307718234e-07,
|
| 1674 |
+
"loss": 0.9159,
|
| 1675 |
+
"step": 23800
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.9861363261264235,
|
| 1679 |
+
"grad_norm": 1.0629512071609497,
|
| 1680 |
+
"learning_rate": 7.077856420626896e-07,
|
| 1681 |
+
"loss": 0.9073,
|
| 1682 |
+
"step": 23900
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.9902624195411784,
|
| 1686 |
+
"grad_norm": 1.0977869033813477,
|
| 1687 |
+
"learning_rate": 4.971351533535558e-07,
|
| 1688 |
+
"loss": 0.9178,
|
| 1689 |
+
"step": 24000
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.9943885129559333,
|
| 1693 |
+
"grad_norm": 1.281014323234558,
|
| 1694 |
+
"learning_rate": 2.8648466464442196e-07,
|
| 1695 |
+
"loss": 0.8876,
|
| 1696 |
+
"step": 24100
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.9985146063706882,
|
| 1700 |
+
"grad_norm": 1.2343029975891113,
|
| 1701 |
+
"learning_rate": 7.583417593528817e-08,
|
| 1702 |
+
"loss": 0.8733,
|
| 1703 |
+
"step": 24200
|
| 1704 |
+
}
|
| 1705 |
+
],
|
| 1706 |
+
"logging_steps": 100,
|
| 1707 |
+
"max_steps": 24236,
|
| 1708 |
+
"num_input_tokens_seen": 0,
|
| 1709 |
+
"num_train_epochs": 1,
|
| 1710 |
+
"save_steps": 10000,
|
| 1711 |
+
"stateful_callbacks": {
|
| 1712 |
+
"TrainerControl": {
|
| 1713 |
+
"args": {
|
| 1714 |
+
"should_epoch_stop": false,
|
| 1715 |
+
"should_evaluate": false,
|
| 1716 |
+
"should_log": false,
|
| 1717 |
+
"should_save": true,
|
| 1718 |
+
"should_training_stop": true
|
| 1719 |
+
},
|
| 1720 |
+
"attributes": {}
|
| 1721 |
+
}
|
| 1722 |
+
},
|
| 1723 |
+
"total_flos": 7.002725362748621e+16,
|
| 1724 |
+
"train_batch_size": 8,
|
| 1725 |
+
"trial_name": null,
|
| 1726 |
+
"trial_params": null
|
| 1727 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0716aa5a07f2f087ae769e110b768f98a4b3d9281c9d033ccb6d0697d199ff8e
|
| 3 |
+
size 5112
|