Spaces:
Sleeping
Sleeping
add check point
Browse files- app.py +2 -2
- checkpoint-2391/config.json +39 -0
- checkpoint-2391/model.safetensors +3 -0
- checkpoint-2391/optimizer.pt +3 -0
- checkpoint-2391/rng_state.pth +3 -0
- checkpoint-2391/scheduler.pt +3 -0
- checkpoint-2391/trainer_state.json +1707 -0
- checkpoint-2391/training_args.bin +3 -0
app.py
CHANGED
|
@@ -4,9 +4,9 @@ import torch
|
|
| 4 |
|
| 5 |
# Load from local checkpoint
|
| 6 |
# or whatever your checkpoint number is
|
| 7 |
-
model_id = "
|
| 8 |
tokenizer = AutoTokenizer.from_pretrained(
|
| 9 |
-
'
|
| 10 |
model = AutoModelForSequenceClassification.from_pretrained(model_id)
|
| 11 |
|
| 12 |
|
|
|
|
| 4 |
|
| 5 |
# Load from local checkpoint
|
| 6 |
# or whatever your checkpoint number is
|
| 7 |
+
model_id = "checkpoint-2391"
|
| 8 |
tokenizer = AutoTokenizer.from_pretrained(
|
| 9 |
+
'huawei-noah/TinyBERT_General_4L_312D') # Original tokenizer
|
| 10 |
model = AutoModelForSequenceClassification.from_pretrained(model_id)
|
| 11 |
|
| 12 |
|
checkpoint-2391/config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"cell": {},
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"emb_size": 312,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 312,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "LABEL_0",
|
| 14 |
+
"1": "LABEL_1",
|
| 15 |
+
"2": "LABEL_2"
|
| 16 |
+
},
|
| 17 |
+
"initializer_range": 0.02,
|
| 18 |
+
"intermediate_size": 1200,
|
| 19 |
+
"label2id": {
|
| 20 |
+
"LABEL_0": 0,
|
| 21 |
+
"LABEL_1": 1,
|
| 22 |
+
"LABEL_2": 2
|
| 23 |
+
},
|
| 24 |
+
"layer_norm_eps": 1e-12,
|
| 25 |
+
"max_position_embeddings": 512,
|
| 26 |
+
"model_type": "bert",
|
| 27 |
+
"num_attention_heads": 12,
|
| 28 |
+
"num_hidden_layers": 4,
|
| 29 |
+
"pad_token_id": 0,
|
| 30 |
+
"position_embedding_type": "absolute",
|
| 31 |
+
"pre_trained": "",
|
| 32 |
+
"problem_type": "single_label_classification",
|
| 33 |
+
"structure": [],
|
| 34 |
+
"torch_dtype": "float32",
|
| 35 |
+
"transformers_version": "4.51.1",
|
| 36 |
+
"type_vocab_size": 2,
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 30522
|
| 39 |
+
}
|
checkpoint-2391/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:99533eb598cf096664292494accc60db13024eeb985d22ade441b1f9420b0960
|
| 3 |
+
size 57413060
|
checkpoint-2391/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9eb8fdbc0461b4c3707aa5b863f38712e4a9d4bf32872cfeea42ba907153b457
|
| 3 |
+
size 114866298
|
checkpoint-2391/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3142eb9ef29f39c1056feaba012d70697a15cdef061694ac7929853189d9ee94
|
| 3 |
+
size 13990
|
checkpoint-2391/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:24a3f9353eec34bf3fee5af0a7ec74bd1c2bcdcb97721880ff38c640da54e774
|
| 3 |
+
size 1064
|
checkpoint-2391/trainer_state.json
ADDED
|
@@ -0,0 +1,1707 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2391,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.012547051442910916,
|
| 14 |
+
"grad_norm": 0.716044008731842,
|
| 15 |
+
"learning_rate": 1.9924717691342537e-05,
|
| 16 |
+
"loss": 1.0972,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.025094102885821833,
|
| 21 |
+
"grad_norm": 1.4613193273544312,
|
| 22 |
+
"learning_rate": 1.9841070681723128e-05,
|
| 23 |
+
"loss": 1.088,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.037641154328732745,
|
| 28 |
+
"grad_norm": 0.627024233341217,
|
| 29 |
+
"learning_rate": 1.9757423672103725e-05,
|
| 30 |
+
"loss": 1.0716,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.050188205771643665,
|
| 35 |
+
"grad_norm": 1.224279522895813,
|
| 36 |
+
"learning_rate": 1.9673776662484318e-05,
|
| 37 |
+
"loss": 1.0661,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.06273525721455459,
|
| 42 |
+
"grad_norm": 1.368181586265564,
|
| 43 |
+
"learning_rate": 1.959012965286491e-05,
|
| 44 |
+
"loss": 1.035,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.07528230865746549,
|
| 49 |
+
"grad_norm": 1.0884851217269897,
|
| 50 |
+
"learning_rate": 1.9506482643245505e-05,
|
| 51 |
+
"loss": 1.0288,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.08782936010037641,
|
| 56 |
+
"grad_norm": 1.9844820499420166,
|
| 57 |
+
"learning_rate": 1.9422835633626102e-05,
|
| 58 |
+
"loss": 0.9855,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.10037641154328733,
|
| 63 |
+
"grad_norm": 1.6572856903076172,
|
| 64 |
+
"learning_rate": 1.9339188624006692e-05,
|
| 65 |
+
"loss": 0.991,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.11292346298619825,
|
| 70 |
+
"grad_norm": 2.2602856159210205,
|
| 71 |
+
"learning_rate": 1.925554161438729e-05,
|
| 72 |
+
"loss": 1.0092,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.12547051442910917,
|
| 77 |
+
"grad_norm": 4.111606597900391,
|
| 78 |
+
"learning_rate": 1.9171894604767882e-05,
|
| 79 |
+
"loss": 0.9062,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.13801756587202008,
|
| 84 |
+
"grad_norm": 3.383659601211548,
|
| 85 |
+
"learning_rate": 1.9088247595148476e-05,
|
| 86 |
+
"loss": 0.9749,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.15056461731493098,
|
| 91 |
+
"grad_norm": 1.9777194261550903,
|
| 92 |
+
"learning_rate": 1.900460058552907e-05,
|
| 93 |
+
"loss": 0.9193,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.16311166875784192,
|
| 98 |
+
"grad_norm": 6.869212627410889,
|
| 99 |
+
"learning_rate": 1.8920953575909663e-05,
|
| 100 |
+
"loss": 0.9695,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.17565872020075282,
|
| 105 |
+
"grad_norm": 2.800828695297241,
|
| 106 |
+
"learning_rate": 1.8837306566290257e-05,
|
| 107 |
+
"loss": 0.9375,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.18820577164366373,
|
| 112 |
+
"grad_norm": 3.783500909805298,
|
| 113 |
+
"learning_rate": 1.875365955667085e-05,
|
| 114 |
+
"loss": 0.9294,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.20075282308657466,
|
| 119 |
+
"grad_norm": 2.961916208267212,
|
| 120 |
+
"learning_rate": 1.8670012547051444e-05,
|
| 121 |
+
"loss": 0.9651,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.21329987452948557,
|
| 126 |
+
"grad_norm": 4.599372386932373,
|
| 127 |
+
"learning_rate": 1.8586365537432037e-05,
|
| 128 |
+
"loss": 0.9706,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.2258469259723965,
|
| 133 |
+
"grad_norm": 3.3036141395568848,
|
| 134 |
+
"learning_rate": 1.8502718527812634e-05,
|
| 135 |
+
"loss": 0.9289,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.2383939774153074,
|
| 140 |
+
"grad_norm": 6.022705078125,
|
| 141 |
+
"learning_rate": 1.8419071518193224e-05,
|
| 142 |
+
"loss": 0.9015,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.25094102885821834,
|
| 147 |
+
"grad_norm": 13.482539176940918,
|
| 148 |
+
"learning_rate": 1.833542450857382e-05,
|
| 149 |
+
"loss": 0.9033,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.26348808030112925,
|
| 154 |
+
"grad_norm": 4.047461032867432,
|
| 155 |
+
"learning_rate": 1.8251777498954414e-05,
|
| 156 |
+
"loss": 0.8447,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.27603513174404015,
|
| 161 |
+
"grad_norm": 2.8806040287017822,
|
| 162 |
+
"learning_rate": 1.8168130489335008e-05,
|
| 163 |
+
"loss": 0.8333,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.28858218318695106,
|
| 168 |
+
"grad_norm": 5.517472743988037,
|
| 169 |
+
"learning_rate": 1.80844834797156e-05,
|
| 170 |
+
"loss": 0.8697,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.30112923462986196,
|
| 175 |
+
"grad_norm": 7.646421432495117,
|
| 176 |
+
"learning_rate": 1.8000836470096195e-05,
|
| 177 |
+
"loss": 0.9159,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.3136762860727729,
|
| 182 |
+
"grad_norm": 4.211366176605225,
|
| 183 |
+
"learning_rate": 1.791718946047679e-05,
|
| 184 |
+
"loss": 0.8848,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.32622333751568383,
|
| 189 |
+
"grad_norm": 7.600694179534912,
|
| 190 |
+
"learning_rate": 1.7833542450857385e-05,
|
| 191 |
+
"loss": 0.8085,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.33877038895859474,
|
| 196 |
+
"grad_norm": 4.872680187225342,
|
| 197 |
+
"learning_rate": 1.7749895441237976e-05,
|
| 198 |
+
"loss": 0.8301,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.35131744040150564,
|
| 203 |
+
"grad_norm": 6.380251884460449,
|
| 204 |
+
"learning_rate": 1.7666248431618572e-05,
|
| 205 |
+
"loss": 0.9484,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.36386449184441655,
|
| 210 |
+
"grad_norm": 4.365920066833496,
|
| 211 |
+
"learning_rate": 1.7582601421999166e-05,
|
| 212 |
+
"loss": 0.9457,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.37641154328732745,
|
| 217 |
+
"grad_norm": 3.710908889770508,
|
| 218 |
+
"learning_rate": 1.749895441237976e-05,
|
| 219 |
+
"loss": 0.8773,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.3889585947302384,
|
| 224 |
+
"grad_norm": 4.279157638549805,
|
| 225 |
+
"learning_rate": 1.7415307402760353e-05,
|
| 226 |
+
"loss": 0.8892,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.4015056461731493,
|
| 231 |
+
"grad_norm": 17.142894744873047,
|
| 232 |
+
"learning_rate": 1.7331660393140947e-05,
|
| 233 |
+
"loss": 0.8723,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.41405269761606023,
|
| 238 |
+
"grad_norm": 3.349973440170288,
|
| 239 |
+
"learning_rate": 1.724801338352154e-05,
|
| 240 |
+
"loss": 0.8092,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.42659974905897113,
|
| 245 |
+
"grad_norm": 7.343029022216797,
|
| 246 |
+
"learning_rate": 1.7164366373902134e-05,
|
| 247 |
+
"loss": 0.9051,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.43914680050188204,
|
| 252 |
+
"grad_norm": 3.02512264251709,
|
| 253 |
+
"learning_rate": 1.7080719364282727e-05,
|
| 254 |
+
"loss": 0.8959,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.451693851944793,
|
| 259 |
+
"grad_norm": 13.072802543640137,
|
| 260 |
+
"learning_rate": 1.699707235466332e-05,
|
| 261 |
+
"loss": 0.9339,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.4642409033877039,
|
| 266 |
+
"grad_norm": 3.5314481258392334,
|
| 267 |
+
"learning_rate": 1.6913425345043917e-05,
|
| 268 |
+
"loss": 0.8883,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.4767879548306148,
|
| 273 |
+
"grad_norm": 3.5592713356018066,
|
| 274 |
+
"learning_rate": 1.6829778335424508e-05,
|
| 275 |
+
"loss": 0.9539,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.4893350062735257,
|
| 280 |
+
"grad_norm": 2.2764039039611816,
|
| 281 |
+
"learning_rate": 1.6746131325805104e-05,
|
| 282 |
+
"loss": 0.9251,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.5018820577164367,
|
| 287 |
+
"grad_norm": 2.953474998474121,
|
| 288 |
+
"learning_rate": 1.6662484316185698e-05,
|
| 289 |
+
"loss": 0.9561,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.5144291091593476,
|
| 294 |
+
"grad_norm": 2.250779628753662,
|
| 295 |
+
"learning_rate": 1.657883730656629e-05,
|
| 296 |
+
"loss": 0.9405,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.5269761606022585,
|
| 301 |
+
"grad_norm": 5.86421537399292,
|
| 302 |
+
"learning_rate": 1.6495190296946885e-05,
|
| 303 |
+
"loss": 0.8936,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.5395232120451694,
|
| 308 |
+
"grad_norm": 4.82138729095459,
|
| 309 |
+
"learning_rate": 1.6411543287327482e-05,
|
| 310 |
+
"loss": 0.8553,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.5520702634880803,
|
| 315 |
+
"grad_norm": 3.090364456176758,
|
| 316 |
+
"learning_rate": 1.6327896277708072e-05,
|
| 317 |
+
"loss": 0.9059,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.5646173149309912,
|
| 322 |
+
"grad_norm": 5.363993167877197,
|
| 323 |
+
"learning_rate": 1.624424926808867e-05,
|
| 324 |
+
"loss": 0.9631,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.5771643663739021,
|
| 329 |
+
"grad_norm": 3.4371540546417236,
|
| 330 |
+
"learning_rate": 1.616060225846926e-05,
|
| 331 |
+
"loss": 0.794,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.589711417816813,
|
| 336 |
+
"grad_norm": 4.080657482147217,
|
| 337 |
+
"learning_rate": 1.6076955248849856e-05,
|
| 338 |
+
"loss": 0.9348,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.6022584692597239,
|
| 343 |
+
"grad_norm": 5.351539134979248,
|
| 344 |
+
"learning_rate": 1.599330823923045e-05,
|
| 345 |
+
"loss": 0.9286,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.6148055207026348,
|
| 350 |
+
"grad_norm": 3.6495888233184814,
|
| 351 |
+
"learning_rate": 1.5909661229611043e-05,
|
| 352 |
+
"loss": 0.7525,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.6273525721455459,
|
| 357 |
+
"grad_norm": 4.849205017089844,
|
| 358 |
+
"learning_rate": 1.5826014219991636e-05,
|
| 359 |
+
"loss": 0.9534,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.6398996235884568,
|
| 364 |
+
"grad_norm": 4.923251152038574,
|
| 365 |
+
"learning_rate": 1.574236721037223e-05,
|
| 366 |
+
"loss": 0.9681,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.6524466750313677,
|
| 371 |
+
"grad_norm": 4.8032073974609375,
|
| 372 |
+
"learning_rate": 1.5658720200752824e-05,
|
| 373 |
+
"loss": 0.8674,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.6649937264742786,
|
| 378 |
+
"grad_norm": 6.583972454071045,
|
| 379 |
+
"learning_rate": 1.5575073191133417e-05,
|
| 380 |
+
"loss": 0.8463,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.6775407779171895,
|
| 385 |
+
"grad_norm": 6.276474475860596,
|
| 386 |
+
"learning_rate": 1.5491426181514014e-05,
|
| 387 |
+
"loss": 0.8481,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.6900878293601004,
|
| 392 |
+
"grad_norm": 6.980823040008545,
|
| 393 |
+
"learning_rate": 1.5407779171894604e-05,
|
| 394 |
+
"loss": 0.8441,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.7026348808030113,
|
| 399 |
+
"grad_norm": 5.919472694396973,
|
| 400 |
+
"learning_rate": 1.53241321622752e-05,
|
| 401 |
+
"loss": 0.7705,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.7151819322459222,
|
| 406 |
+
"grad_norm": 14.738715171813965,
|
| 407 |
+
"learning_rate": 1.5240485152655793e-05,
|
| 408 |
+
"loss": 0.8883,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.7277289836888331,
|
| 413 |
+
"grad_norm": 9.944400787353516,
|
| 414 |
+
"learning_rate": 1.5156838143036388e-05,
|
| 415 |
+
"loss": 0.843,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.740276035131744,
|
| 420 |
+
"grad_norm": 5.619649410247803,
|
| 421 |
+
"learning_rate": 1.5073191133416981e-05,
|
| 422 |
+
"loss": 0.9996,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.7528230865746549,
|
| 427 |
+
"grad_norm": 7.122910499572754,
|
| 428 |
+
"learning_rate": 1.4989544123797575e-05,
|
| 429 |
+
"loss": 0.9,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.7653701380175659,
|
| 434 |
+
"grad_norm": 10.766443252563477,
|
| 435 |
+
"learning_rate": 1.490589711417817e-05,
|
| 436 |
+
"loss": 0.8326,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.7779171894604768,
|
| 441 |
+
"grad_norm": 4.103264808654785,
|
| 442 |
+
"learning_rate": 1.4822250104558764e-05,
|
| 443 |
+
"loss": 0.8552,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.7904642409033877,
|
| 448 |
+
"grad_norm": 11.64675521850586,
|
| 449 |
+
"learning_rate": 1.4738603094939357e-05,
|
| 450 |
+
"loss": 0.774,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.8030112923462986,
|
| 455 |
+
"grad_norm": 6.035724639892578,
|
| 456 |
+
"learning_rate": 1.465495608531995e-05,
|
| 457 |
+
"loss": 0.9782,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.8155583437892095,
|
| 462 |
+
"grad_norm": 6.475778102874756,
|
| 463 |
+
"learning_rate": 1.4571309075700546e-05,
|
| 464 |
+
"loss": 0.7997,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.8281053952321205,
|
| 469 |
+
"grad_norm": 5.5077314376831055,
|
| 470 |
+
"learning_rate": 1.4487662066081138e-05,
|
| 471 |
+
"loss": 0.7985,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.8406524466750314,
|
| 476 |
+
"grad_norm": 9.788359642028809,
|
| 477 |
+
"learning_rate": 1.4404015056461733e-05,
|
| 478 |
+
"loss": 0.9237,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.8531994981179423,
|
| 483 |
+
"grad_norm": 5.642941951751709,
|
| 484 |
+
"learning_rate": 1.4320368046842325e-05,
|
| 485 |
+
"loss": 0.7795,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.8657465495608532,
|
| 490 |
+
"grad_norm": 7.9442315101623535,
|
| 491 |
+
"learning_rate": 1.423672103722292e-05,
|
| 492 |
+
"loss": 0.891,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.8782936010037641,
|
| 497 |
+
"grad_norm": 4.023198127746582,
|
| 498 |
+
"learning_rate": 1.4153074027603515e-05,
|
| 499 |
+
"loss": 0.8391,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.890840652446675,
|
| 504 |
+
"grad_norm": 11.08212661743164,
|
| 505 |
+
"learning_rate": 1.4069427017984107e-05,
|
| 506 |
+
"loss": 0.9049,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.903387703889586,
|
| 511 |
+
"grad_norm": 7.001065254211426,
|
| 512 |
+
"learning_rate": 1.3985780008364702e-05,
|
| 513 |
+
"loss": 0.9636,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.9159347553324969,
|
| 518 |
+
"grad_norm": 7.488030433654785,
|
| 519 |
+
"learning_rate": 1.3902132998745297e-05,
|
| 520 |
+
"loss": 0.8491,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.9284818067754078,
|
| 525 |
+
"grad_norm": 5.319072246551514,
|
| 526 |
+
"learning_rate": 1.381848598912589e-05,
|
| 527 |
+
"loss": 0.921,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.9410288582183187,
|
| 532 |
+
"grad_norm": 7.3197126388549805,
|
| 533 |
+
"learning_rate": 1.3734838979506484e-05,
|
| 534 |
+
"loss": 0.9866,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.9535759096612296,
|
| 539 |
+
"grad_norm": 5.280470848083496,
|
| 540 |
+
"learning_rate": 1.3651191969887078e-05,
|
| 541 |
+
"loss": 0.8146,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.9661229611041405,
|
| 546 |
+
"grad_norm": 3.304518938064575,
|
| 547 |
+
"learning_rate": 1.3567544960267671e-05,
|
| 548 |
+
"loss": 0.8724,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.9786700125470514,
|
| 553 |
+
"grad_norm": 3.306471109390259,
|
| 554 |
+
"learning_rate": 1.3483897950648267e-05,
|
| 555 |
+
"loss": 0.8796,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.9912170639899623,
|
| 560 |
+
"grad_norm": 5.108331203460693,
|
| 561 |
+
"learning_rate": 1.3400250941028858e-05,
|
| 562 |
+
"loss": 0.8931,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 1.0037641154328734,
|
| 567 |
+
"grad_norm": 9.46453857421875,
|
| 568 |
+
"learning_rate": 1.3316603931409454e-05,
|
| 569 |
+
"loss": 0.8834,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.0163111668757843,
|
| 574 |
+
"grad_norm": 8.54848861694336,
|
| 575 |
+
"learning_rate": 1.3232956921790047e-05,
|
| 576 |
+
"loss": 0.7073,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 1.0288582183186952,
|
| 581 |
+
"grad_norm": 8.689090728759766,
|
| 582 |
+
"learning_rate": 1.314930991217064e-05,
|
| 583 |
+
"loss": 0.8242,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 1.041405269761606,
|
| 588 |
+
"grad_norm": 8.000664710998535,
|
| 589 |
+
"learning_rate": 1.3065662902551234e-05,
|
| 590 |
+
"loss": 0.8314,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.053952321204517,
|
| 595 |
+
"grad_norm": 11.315167427062988,
|
| 596 |
+
"learning_rate": 1.298201589293183e-05,
|
| 597 |
+
"loss": 0.6794,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 1.066499372647428,
|
| 602 |
+
"grad_norm": 6.037744522094727,
|
| 603 |
+
"learning_rate": 1.2898368883312423e-05,
|
| 604 |
+
"loss": 0.7796,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 1.0790464240903388,
|
| 609 |
+
"grad_norm": 9.521272659301758,
|
| 610 |
+
"learning_rate": 1.2814721873693016e-05,
|
| 611 |
+
"loss": 0.92,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 1.0915934755332497,
|
| 616 |
+
"grad_norm": 18.728450775146484,
|
| 617 |
+
"learning_rate": 1.2731074864073612e-05,
|
| 618 |
+
"loss": 0.8371,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 1.1041405269761606,
|
| 623 |
+
"grad_norm": 7.143416404724121,
|
| 624 |
+
"learning_rate": 1.2647427854454203e-05,
|
| 625 |
+
"loss": 0.8534,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 1.1166875784190715,
|
| 630 |
+
"grad_norm": 7.581761837005615,
|
| 631 |
+
"learning_rate": 1.2563780844834799e-05,
|
| 632 |
+
"loss": 0.894,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.1292346298619824,
|
| 637 |
+
"grad_norm": 3.4014008045196533,
|
| 638 |
+
"learning_rate": 1.248013383521539e-05,
|
| 639 |
+
"loss": 0.8155,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.1417816813048933,
|
| 644 |
+
"grad_norm": 13.978635787963867,
|
| 645 |
+
"learning_rate": 1.2396486825595986e-05,
|
| 646 |
+
"loss": 0.754,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.1543287327478042,
|
| 651 |
+
"grad_norm": 5.18642520904541,
|
| 652 |
+
"learning_rate": 1.2312839815976581e-05,
|
| 653 |
+
"loss": 0.8493,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.1668757841907151,
|
| 658 |
+
"grad_norm": 4.153924942016602,
|
| 659 |
+
"learning_rate": 1.2229192806357173e-05,
|
| 660 |
+
"loss": 0.813,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.179422835633626,
|
| 665 |
+
"grad_norm": 9.386292457580566,
|
| 666 |
+
"learning_rate": 1.2145545796737768e-05,
|
| 667 |
+
"loss": 0.7526,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.191969887076537,
|
| 672 |
+
"grad_norm": 5.997525215148926,
|
| 673 |
+
"learning_rate": 1.2061898787118363e-05,
|
| 674 |
+
"loss": 0.8602,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.2045169385194479,
|
| 679 |
+
"grad_norm": 7.475282669067383,
|
| 680 |
+
"learning_rate": 1.1978251777498955e-05,
|
| 681 |
+
"loss": 0.6822,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.2170639899623588,
|
| 686 |
+
"grad_norm": 11.12500286102295,
|
| 687 |
+
"learning_rate": 1.189460476787955e-05,
|
| 688 |
+
"loss": 0.7505,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.2296110414052697,
|
| 693 |
+
"grad_norm": 10.952397346496582,
|
| 694 |
+
"learning_rate": 1.1810957758260144e-05,
|
| 695 |
+
"loss": 0.8621,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.2421580928481806,
|
| 700 |
+
"grad_norm": 16.098909378051758,
|
| 701 |
+
"learning_rate": 1.1727310748640737e-05,
|
| 702 |
+
"loss": 0.8263,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.2547051442910915,
|
| 707 |
+
"grad_norm": 7.630945682525635,
|
| 708 |
+
"learning_rate": 1.164366373902133e-05,
|
| 709 |
+
"loss": 0.7283,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 1.2672521957340024,
|
| 714 |
+
"grad_norm": 5.213379859924316,
|
| 715 |
+
"learning_rate": 1.1560016729401924e-05,
|
| 716 |
+
"loss": 0.7849,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 1.2797992471769133,
|
| 721 |
+
"grad_norm": 9.576950073242188,
|
| 722 |
+
"learning_rate": 1.147636971978252e-05,
|
| 723 |
+
"loss": 0.7845,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 1.2923462986198244,
|
| 728 |
+
"grad_norm": 7.526945114135742,
|
| 729 |
+
"learning_rate": 1.1392722710163113e-05,
|
| 730 |
+
"loss": 0.8226,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 1.3048933500627353,
|
| 735 |
+
"grad_norm": 8.773786544799805,
|
| 736 |
+
"learning_rate": 1.1309075700543706e-05,
|
| 737 |
+
"loss": 0.7912,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 1.3174404015056462,
|
| 742 |
+
"grad_norm": 5.114284038543701,
|
| 743 |
+
"learning_rate": 1.12254286909243e-05,
|
| 744 |
+
"loss": 0.7414,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 1.3299874529485571,
|
| 749 |
+
"grad_norm": 6.488368988037109,
|
| 750 |
+
"learning_rate": 1.1141781681304895e-05,
|
| 751 |
+
"loss": 0.8588,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 1.342534504391468,
|
| 756 |
+
"grad_norm": 6.658433437347412,
|
| 757 |
+
"learning_rate": 1.1058134671685487e-05,
|
| 758 |
+
"loss": 0.9214,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 1.355081555834379,
|
| 763 |
+
"grad_norm": 10.264144897460938,
|
| 764 |
+
"learning_rate": 1.0974487662066082e-05,
|
| 765 |
+
"loss": 0.7492,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 1.3676286072772899,
|
| 770 |
+
"grad_norm": 7.86182165145874,
|
| 771 |
+
"learning_rate": 1.0890840652446677e-05,
|
| 772 |
+
"loss": 0.7919,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 1.3801756587202008,
|
| 777 |
+
"grad_norm": 7.3965535163879395,
|
| 778 |
+
"learning_rate": 1.080719364282727e-05,
|
| 779 |
+
"loss": 0.8695,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 1.3927227101631117,
|
| 784 |
+
"grad_norm": 9.31979751586914,
|
| 785 |
+
"learning_rate": 1.0723546633207864e-05,
|
| 786 |
+
"loss": 0.7495,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 1.4052697616060226,
|
| 791 |
+
"grad_norm": 9.471373558044434,
|
| 792 |
+
"learning_rate": 1.0639899623588456e-05,
|
| 793 |
+
"loss": 0.8729,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.4178168130489335,
|
| 798 |
+
"grad_norm": 13.156733512878418,
|
| 799 |
+
"learning_rate": 1.0556252613969051e-05,
|
| 800 |
+
"loss": 0.8688,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 1.4303638644918444,
|
| 805 |
+
"grad_norm": 13.249898910522461,
|
| 806 |
+
"learning_rate": 1.0472605604349647e-05,
|
| 807 |
+
"loss": 0.8097,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 1.4429109159347553,
|
| 812 |
+
"grad_norm": 6.98345947265625,
|
| 813 |
+
"learning_rate": 1.0388958594730238e-05,
|
| 814 |
+
"loss": 0.7635,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 1.4554579673776662,
|
| 819 |
+
"grad_norm": 18.78076171875,
|
| 820 |
+
"learning_rate": 1.0305311585110834e-05,
|
| 821 |
+
"loss": 0.957,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 1.468005018820577,
|
| 826 |
+
"grad_norm": 25.165569305419922,
|
| 827 |
+
"learning_rate": 1.0221664575491427e-05,
|
| 828 |
+
"loss": 0.7616,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 1.480552070263488,
|
| 833 |
+
"grad_norm": 8.38501262664795,
|
| 834 |
+
"learning_rate": 1.013801756587202e-05,
|
| 835 |
+
"loss": 0.7768,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 1.4930991217063991,
|
| 840 |
+
"grad_norm": 10.194948196411133,
|
| 841 |
+
"learning_rate": 1.0054370556252616e-05,
|
| 842 |
+
"loss": 0.8924,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 1.50564617314931,
|
| 847 |
+
"grad_norm": 10.704676628112793,
|
| 848 |
+
"learning_rate": 9.97072354663321e-06,
|
| 849 |
+
"loss": 0.9066,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 1.518193224592221,
|
| 854 |
+
"grad_norm": 21.507707595825195,
|
| 855 |
+
"learning_rate": 9.887076537013803e-06,
|
| 856 |
+
"loss": 0.9013,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 1.5307402760351319,
|
| 861 |
+
"grad_norm": 8.45915699005127,
|
| 862 |
+
"learning_rate": 9.803429527394396e-06,
|
| 863 |
+
"loss": 1.0474,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 1.5432873274780428,
|
| 868 |
+
"grad_norm": 5.320367336273193,
|
| 869 |
+
"learning_rate": 9.71978251777499e-06,
|
| 870 |
+
"loss": 0.8313,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 1.5558343789209537,
|
| 875 |
+
"grad_norm": 10.026762962341309,
|
| 876 |
+
"learning_rate": 9.636135508155583e-06,
|
| 877 |
+
"loss": 0.8881,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 1.5683814303638646,
|
| 882 |
+
"grad_norm": 9.318647384643555,
|
| 883 |
+
"learning_rate": 9.552488498536177e-06,
|
| 884 |
+
"loss": 0.8274,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 1.5809284818067755,
|
| 889 |
+
"grad_norm": 9.438924789428711,
|
| 890 |
+
"learning_rate": 9.468841488916772e-06,
|
| 891 |
+
"loss": 0.9072,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 1.5934755332496864,
|
| 896 |
+
"grad_norm": 11.8627290725708,
|
| 897 |
+
"learning_rate": 9.385194479297366e-06,
|
| 898 |
+
"loss": 0.746,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 1.6060225846925973,
|
| 903 |
+
"grad_norm": 5.647990703582764,
|
| 904 |
+
"learning_rate": 9.30154746967796e-06,
|
| 905 |
+
"loss": 0.8526,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 1.6185696361355082,
|
| 910 |
+
"grad_norm": 17.41633415222168,
|
| 911 |
+
"learning_rate": 9.217900460058554e-06,
|
| 912 |
+
"loss": 0.7216,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 1.631116687578419,
|
| 917 |
+
"grad_norm": 10.73770523071289,
|
| 918 |
+
"learning_rate": 9.134253450439148e-06,
|
| 919 |
+
"loss": 0.7405,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 1.64366373902133,
|
| 924 |
+
"grad_norm": 9.316720962524414,
|
| 925 |
+
"learning_rate": 9.050606440819741e-06,
|
| 926 |
+
"loss": 0.727,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 1.656210790464241,
|
| 931 |
+
"grad_norm": 8.067049980163574,
|
| 932 |
+
"learning_rate": 8.966959431200335e-06,
|
| 933 |
+
"loss": 0.7992,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 1.6687578419071518,
|
| 938 |
+
"grad_norm": 23.564279556274414,
|
| 939 |
+
"learning_rate": 8.88331242158093e-06,
|
| 940 |
+
"loss": 0.7617,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.6813048933500627,
|
| 945 |
+
"grad_norm": 10.805792808532715,
|
| 946 |
+
"learning_rate": 8.799665411961524e-06,
|
| 947 |
+
"loss": 0.9413,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.6938519447929736,
|
| 952 |
+
"grad_norm": 8.173852920532227,
|
| 953 |
+
"learning_rate": 8.716018402342117e-06,
|
| 954 |
+
"loss": 0.8204,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.7063989962358845,
|
| 959 |
+
"grad_norm": 16.444934844970703,
|
| 960 |
+
"learning_rate": 8.63237139272271e-06,
|
| 961 |
+
"loss": 0.7124,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.7189460476787954,
|
| 966 |
+
"grad_norm": 19.09043312072754,
|
| 967 |
+
"learning_rate": 8.548724383103306e-06,
|
| 968 |
+
"loss": 0.8573,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.7314930991217063,
|
| 973 |
+
"grad_norm": 15.454293251037598,
|
| 974 |
+
"learning_rate": 8.4650773734839e-06,
|
| 975 |
+
"loss": 0.7964,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.7440401505646173,
|
| 980 |
+
"grad_norm": 9.219696044921875,
|
| 981 |
+
"learning_rate": 8.381430363864493e-06,
|
| 982 |
+
"loss": 0.7494,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.7565872020075282,
|
| 987 |
+
"grad_norm": 12.434812545776367,
|
| 988 |
+
"learning_rate": 8.297783354245086e-06,
|
| 989 |
+
"loss": 0.7066,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.769134253450439,
|
| 994 |
+
"grad_norm": 8.64736270904541,
|
| 995 |
+
"learning_rate": 8.21413634462568e-06,
|
| 996 |
+
"loss": 0.8049,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.78168130489335,
|
| 1001 |
+
"grad_norm": 8.343708038330078,
|
| 1002 |
+
"learning_rate": 8.130489335006273e-06,
|
| 1003 |
+
"loss": 0.7935,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.7942283563362609,
|
| 1008 |
+
"grad_norm": 6.48344612121582,
|
| 1009 |
+
"learning_rate": 8.046842325386867e-06,
|
| 1010 |
+
"loss": 0.7431,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.8067754077791718,
|
| 1015 |
+
"grad_norm": 7.362094402313232,
|
| 1016 |
+
"learning_rate": 7.963195315767462e-06,
|
| 1017 |
+
"loss": 0.6778,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.8193224592220827,
|
| 1022 |
+
"grad_norm": 10.41412353515625,
|
| 1023 |
+
"learning_rate": 7.879548306148056e-06,
|
| 1024 |
+
"loss": 0.8869,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.8318695106649936,
|
| 1029 |
+
"grad_norm": 11.075960159301758,
|
| 1030 |
+
"learning_rate": 7.79590129652865e-06,
|
| 1031 |
+
"loss": 0.8913,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.8444165621079045,
|
| 1036 |
+
"grad_norm": 3.257545232772827,
|
| 1037 |
+
"learning_rate": 7.712254286909243e-06,
|
| 1038 |
+
"loss": 0.8214,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.8569636135508154,
|
| 1043 |
+
"grad_norm": 2.363823175430298,
|
| 1044 |
+
"learning_rate": 7.628607277289838e-06,
|
| 1045 |
+
"loss": 0.8268,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.8695106649937263,
|
| 1050 |
+
"grad_norm": 16.772573471069336,
|
| 1051 |
+
"learning_rate": 7.544960267670431e-06,
|
| 1052 |
+
"loss": 0.9585,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.8820577164366374,
|
| 1057 |
+
"grad_norm": 23.401941299438477,
|
| 1058 |
+
"learning_rate": 7.461313258051025e-06,
|
| 1059 |
+
"loss": 0.8636,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.8946047678795483,
|
| 1064 |
+
"grad_norm": 8.009302139282227,
|
| 1065 |
+
"learning_rate": 7.37766624843162e-06,
|
| 1066 |
+
"loss": 0.6728,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.9071518193224593,
|
| 1071 |
+
"grad_norm": 19.989112854003906,
|
| 1072 |
+
"learning_rate": 7.294019238812214e-06,
|
| 1073 |
+
"loss": 0.8526,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.9196988707653702,
|
| 1078 |
+
"grad_norm": 11.849247932434082,
|
| 1079 |
+
"learning_rate": 7.210372229192807e-06,
|
| 1080 |
+
"loss": 0.8825,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.932245922208281,
|
| 1085 |
+
"grad_norm": 8.389686584472656,
|
| 1086 |
+
"learning_rate": 7.126725219573401e-06,
|
| 1087 |
+
"loss": 0.8261,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.944792973651192,
|
| 1092 |
+
"grad_norm": 11.961833000183105,
|
| 1093 |
+
"learning_rate": 7.043078209953995e-06,
|
| 1094 |
+
"loss": 0.6564,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.9573400250941029,
|
| 1099 |
+
"grad_norm": 15.514620780944824,
|
| 1100 |
+
"learning_rate": 6.9594312003345885e-06,
|
| 1101 |
+
"loss": 0.7943,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.9698870765370138,
|
| 1106 |
+
"grad_norm": 10.584662437438965,
|
| 1107 |
+
"learning_rate": 6.875784190715182e-06,
|
| 1108 |
+
"loss": 0.7593,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.9824341279799247,
|
| 1113 |
+
"grad_norm": 15.600502014160156,
|
| 1114 |
+
"learning_rate": 6.7921371810957755e-06,
|
| 1115 |
+
"loss": 0.8643,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.9949811794228356,
|
| 1120 |
+
"grad_norm": 11.883516311645508,
|
| 1121 |
+
"learning_rate": 6.708490171476371e-06,
|
| 1122 |
+
"loss": 0.8381,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 2.0075282308657467,
|
| 1127 |
+
"grad_norm": 6.613313674926758,
|
| 1128 |
+
"learning_rate": 6.624843161856964e-06,
|
| 1129 |
+
"loss": 0.8074,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 2.0200752823086576,
|
| 1134 |
+
"grad_norm": 10.316112518310547,
|
| 1135 |
+
"learning_rate": 6.541196152237558e-06,
|
| 1136 |
+
"loss": 0.7161,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 2.0326223337515685,
|
| 1141 |
+
"grad_norm": 34.29435348510742,
|
| 1142 |
+
"learning_rate": 6.457549142618152e-06,
|
| 1143 |
+
"loss": 0.7553,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 2.0451693851944794,
|
| 1148 |
+
"grad_norm": 13.848133087158203,
|
| 1149 |
+
"learning_rate": 6.373902132998746e-06,
|
| 1150 |
+
"loss": 0.7486,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 2.0577164366373903,
|
| 1155 |
+
"grad_norm": 6.787230014801025,
|
| 1156 |
+
"learning_rate": 6.29025512337934e-06,
|
| 1157 |
+
"loss": 0.7867,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 2.0702634880803013,
|
| 1162 |
+
"grad_norm": 12.09325122833252,
|
| 1163 |
+
"learning_rate": 6.2066081137599335e-06,
|
| 1164 |
+
"loss": 0.7999,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 2.082810539523212,
|
| 1169 |
+
"grad_norm": 9.453817367553711,
|
| 1170 |
+
"learning_rate": 6.122961104140528e-06,
|
| 1171 |
+
"loss": 0.7668,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 2.095357590966123,
|
| 1176 |
+
"grad_norm": 10.155051231384277,
|
| 1177 |
+
"learning_rate": 6.039314094521121e-06,
|
| 1178 |
+
"loss": 0.7595,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 2.107904642409034,
|
| 1183 |
+
"grad_norm": 26.19032859802246,
|
| 1184 |
+
"learning_rate": 5.955667084901715e-06,
|
| 1185 |
+
"loss": 0.9065,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 2.120451693851945,
|
| 1190 |
+
"grad_norm": 9.538314819335938,
|
| 1191 |
+
"learning_rate": 5.872020075282308e-06,
|
| 1192 |
+
"loss": 0.7759,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 2.132998745294856,
|
| 1197 |
+
"grad_norm": 6.306014060974121,
|
| 1198 |
+
"learning_rate": 5.788373065662904e-06,
|
| 1199 |
+
"loss": 0.7431,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 2.1455457967377667,
|
| 1204 |
+
"grad_norm": 8.156487464904785,
|
| 1205 |
+
"learning_rate": 5.704726056043497e-06,
|
| 1206 |
+
"loss": 0.7029,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 2.1580928481806776,
|
| 1211 |
+
"grad_norm": 14.053692817687988,
|
| 1212 |
+
"learning_rate": 5.621079046424091e-06,
|
| 1213 |
+
"loss": 0.8357,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 2.1706398996235885,
|
| 1218 |
+
"grad_norm": 10.932027816772461,
|
| 1219 |
+
"learning_rate": 5.537432036804685e-06,
|
| 1220 |
+
"loss": 0.723,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 2.1831869510664994,
|
| 1225 |
+
"grad_norm": 23.12853240966797,
|
| 1226 |
+
"learning_rate": 5.4537850271852785e-06,
|
| 1227 |
+
"loss": 0.819,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 2.1957340025094103,
|
| 1232 |
+
"grad_norm": 10.468832015991211,
|
| 1233 |
+
"learning_rate": 5.370138017565872e-06,
|
| 1234 |
+
"loss": 0.7771,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 2.208281053952321,
|
| 1239 |
+
"grad_norm": 16.84822654724121,
|
| 1240 |
+
"learning_rate": 5.286491007946466e-06,
|
| 1241 |
+
"loss": 0.7895,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 2.220828105395232,
|
| 1246 |
+
"grad_norm": 17.586076736450195,
|
| 1247 |
+
"learning_rate": 5.202843998327061e-06,
|
| 1248 |
+
"loss": 0.8417,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 2.233375156838143,
|
| 1253 |
+
"grad_norm": 5.2970662117004395,
|
| 1254 |
+
"learning_rate": 5.119196988707654e-06,
|
| 1255 |
+
"loss": 0.7451,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 2.245922208281054,
|
| 1260 |
+
"grad_norm": 11.92138385772705,
|
| 1261 |
+
"learning_rate": 5.035549979088248e-06,
|
| 1262 |
+
"loss": 0.8368,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 2.258469259723965,
|
| 1267 |
+
"grad_norm": 22.60080909729004,
|
| 1268 |
+
"learning_rate": 4.951902969468842e-06,
|
| 1269 |
+
"loss": 0.7133,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 2.2710163111668757,
|
| 1274 |
+
"grad_norm": 12.277490615844727,
|
| 1275 |
+
"learning_rate": 4.868255959849436e-06,
|
| 1276 |
+
"loss": 0.8418,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 2.2835633626097867,
|
| 1281 |
+
"grad_norm": 6.862631320953369,
|
| 1282 |
+
"learning_rate": 4.78460895023003e-06,
|
| 1283 |
+
"loss": 0.7564,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 2.2961104140526976,
|
| 1288 |
+
"grad_norm": 7.9091267585754395,
|
| 1289 |
+
"learning_rate": 4.7009619406106235e-06,
|
| 1290 |
+
"loss": 0.7582,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 2.3086574654956085,
|
| 1295 |
+
"grad_norm": 8.295835494995117,
|
| 1296 |
+
"learning_rate": 4.617314930991217e-06,
|
| 1297 |
+
"loss": 0.8465,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 2.3212045169385194,
|
| 1302 |
+
"grad_norm": 11.34477710723877,
|
| 1303 |
+
"learning_rate": 4.533667921371811e-06,
|
| 1304 |
+
"loss": 0.7846,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 2.3337515683814303,
|
| 1309 |
+
"grad_norm": 32.95912170410156,
|
| 1310 |
+
"learning_rate": 4.450020911752405e-06,
|
| 1311 |
+
"loss": 0.6656,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 2.346298619824341,
|
| 1316 |
+
"grad_norm": 17.355579376220703,
|
| 1317 |
+
"learning_rate": 4.366373902132999e-06,
|
| 1318 |
+
"loss": 0.8187,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 2.358845671267252,
|
| 1323 |
+
"grad_norm": 8.741580963134766,
|
| 1324 |
+
"learning_rate": 4.282726892513593e-06,
|
| 1325 |
+
"loss": 0.744,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 2.371392722710163,
|
| 1330 |
+
"grad_norm": 12.800196647644043,
|
| 1331 |
+
"learning_rate": 4.199079882894187e-06,
|
| 1332 |
+
"loss": 0.6891,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 2.383939774153074,
|
| 1337 |
+
"grad_norm": 6.387624740600586,
|
| 1338 |
+
"learning_rate": 4.115432873274781e-06,
|
| 1339 |
+
"loss": 0.6968,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 2.396486825595985,
|
| 1344 |
+
"grad_norm": 9.609091758728027,
|
| 1345 |
+
"learning_rate": 4.031785863655375e-06,
|
| 1346 |
+
"loss": 0.7412,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 2.4090338770388957,
|
| 1351 |
+
"grad_norm": 16.369497299194336,
|
| 1352 |
+
"learning_rate": 3.9481388540359685e-06,
|
| 1353 |
+
"loss": 0.7136,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 2.4215809284818066,
|
| 1358 |
+
"grad_norm": 7.245341777801514,
|
| 1359 |
+
"learning_rate": 3.864491844416563e-06,
|
| 1360 |
+
"loss": 0.6953,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 2.4341279799247175,
|
| 1365 |
+
"grad_norm": 21.487659454345703,
|
| 1366 |
+
"learning_rate": 3.7808448347971568e-06,
|
| 1367 |
+
"loss": 0.776,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 2.4466750313676284,
|
| 1372 |
+
"grad_norm": 13.91763973236084,
|
| 1373 |
+
"learning_rate": 3.6971978251777503e-06,
|
| 1374 |
+
"loss": 0.9644,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 2.4592220828105393,
|
| 1379 |
+
"grad_norm": 16.034120559692383,
|
| 1380 |
+
"learning_rate": 3.6135508155583442e-06,
|
| 1381 |
+
"loss": 0.7451,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 2.4717691342534502,
|
| 1386 |
+
"grad_norm": 15.478806495666504,
|
| 1387 |
+
"learning_rate": 3.5299038059389377e-06,
|
| 1388 |
+
"loss": 0.6669,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 2.484316185696361,
|
| 1393 |
+
"grad_norm": 14.39084529876709,
|
| 1394 |
+
"learning_rate": 3.446256796319532e-06,
|
| 1395 |
+
"loss": 0.7545,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 2.496863237139272,
|
| 1400 |
+
"grad_norm": 27.011598587036133,
|
| 1401 |
+
"learning_rate": 3.3626097867001256e-06,
|
| 1402 |
+
"loss": 0.8599,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 2.509410288582183,
|
| 1407 |
+
"grad_norm": 15.559931755065918,
|
| 1408 |
+
"learning_rate": 3.27896277708072e-06,
|
| 1409 |
+
"loss": 0.8266,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 2.521957340025094,
|
| 1414 |
+
"grad_norm": 8.093729019165039,
|
| 1415 |
+
"learning_rate": 3.1953157674613135e-06,
|
| 1416 |
+
"loss": 0.7978,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 2.5345043914680048,
|
| 1421 |
+
"grad_norm": 13.65913200378418,
|
| 1422 |
+
"learning_rate": 3.1116687578419074e-06,
|
| 1423 |
+
"loss": 0.8145,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 2.5470514429109157,
|
| 1428 |
+
"grad_norm": 10.85305404663086,
|
| 1429 |
+
"learning_rate": 3.028021748222501e-06,
|
| 1430 |
+
"loss": 0.7976,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 2.5595984943538266,
|
| 1435 |
+
"grad_norm": 21.572189331054688,
|
| 1436 |
+
"learning_rate": 2.9443747386030953e-06,
|
| 1437 |
+
"loss": 0.7435,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 2.572145545796738,
|
| 1442 |
+
"grad_norm": 14.962433815002441,
|
| 1443 |
+
"learning_rate": 2.860727728983689e-06,
|
| 1444 |
+
"loss": 0.7856,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 2.584692597239649,
|
| 1449 |
+
"grad_norm": 8.81922435760498,
|
| 1450 |
+
"learning_rate": 2.7770807193642827e-06,
|
| 1451 |
+
"loss": 0.6729,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 2.5972396486825597,
|
| 1456 |
+
"grad_norm": 36.92323684692383,
|
| 1457 |
+
"learning_rate": 2.693433709744877e-06,
|
| 1458 |
+
"loss": 0.7374,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 2.6097867001254706,
|
| 1463 |
+
"grad_norm": 19.693674087524414,
|
| 1464 |
+
"learning_rate": 2.6097867001254706e-06,
|
| 1465 |
+
"loss": 0.6577,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 2.6223337515683816,
|
| 1470 |
+
"grad_norm": 6.363992691040039,
|
| 1471 |
+
"learning_rate": 2.526139690506065e-06,
|
| 1472 |
+
"loss": 0.7902,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 2.6348808030112925,
|
| 1477 |
+
"grad_norm": 18.565412521362305,
|
| 1478 |
+
"learning_rate": 2.4424926808866585e-06,
|
| 1479 |
+
"loss": 0.638,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 2.6474278544542034,
|
| 1484 |
+
"grad_norm": 23.991640090942383,
|
| 1485 |
+
"learning_rate": 2.3588456712672524e-06,
|
| 1486 |
+
"loss": 0.8057,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 2.6599749058971143,
|
| 1491 |
+
"grad_norm": 11.251315116882324,
|
| 1492 |
+
"learning_rate": 2.2751986616478463e-06,
|
| 1493 |
+
"loss": 0.7766,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 2.672521957340025,
|
| 1498 |
+
"grad_norm": 13.534173011779785,
|
| 1499 |
+
"learning_rate": 2.1915516520284403e-06,
|
| 1500 |
+
"loss": 0.7792,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 2.685069008782936,
|
| 1505 |
+
"grad_norm": 28.147571563720703,
|
| 1506 |
+
"learning_rate": 2.107904642409034e-06,
|
| 1507 |
+
"loss": 0.7725,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 2.697616060225847,
|
| 1512 |
+
"grad_norm": 25.395660400390625,
|
| 1513 |
+
"learning_rate": 2.024257632789628e-06,
|
| 1514 |
+
"loss": 0.7114,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 2.710163111668758,
|
| 1519 |
+
"grad_norm": 11.658987998962402,
|
| 1520 |
+
"learning_rate": 1.9406106231702216e-06,
|
| 1521 |
+
"loss": 0.7515,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 2.722710163111669,
|
| 1526 |
+
"grad_norm": 7.770451068878174,
|
| 1527 |
+
"learning_rate": 1.8569636135508156e-06,
|
| 1528 |
+
"loss": 0.7042,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 2.7352572145545797,
|
| 1533 |
+
"grad_norm": 15.549407005310059,
|
| 1534 |
+
"learning_rate": 1.7733166039314095e-06,
|
| 1535 |
+
"loss": 0.8029,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 2.7478042659974906,
|
| 1540 |
+
"grad_norm": 9.49413013458252,
|
| 1541 |
+
"learning_rate": 1.6896695943120034e-06,
|
| 1542 |
+
"loss": 0.642,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 2.7603513174404015,
|
| 1547 |
+
"grad_norm": 4.831557750701904,
|
| 1548 |
+
"learning_rate": 1.6060225846925972e-06,
|
| 1549 |
+
"loss": 0.6564,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 2.7728983688833124,
|
| 1554 |
+
"grad_norm": 12.240572929382324,
|
| 1555 |
+
"learning_rate": 1.5223755750731913e-06,
|
| 1556 |
+
"loss": 0.7423,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 2.7854454203262233,
|
| 1561 |
+
"grad_norm": 5.629863739013672,
|
| 1562 |
+
"learning_rate": 1.4387285654537853e-06,
|
| 1563 |
+
"loss": 0.8664,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 2.7979924717691342,
|
| 1568 |
+
"grad_norm": 14.666424751281738,
|
| 1569 |
+
"learning_rate": 1.3550815558343792e-06,
|
| 1570 |
+
"loss": 0.8401,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 2.810539523212045,
|
| 1575 |
+
"grad_norm": 21.368179321289062,
|
| 1576 |
+
"learning_rate": 1.271434546214973e-06,
|
| 1577 |
+
"loss": 0.8129,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 2.823086574654956,
|
| 1582 |
+
"grad_norm": 16.279918670654297,
|
| 1583 |
+
"learning_rate": 1.1877875365955668e-06,
|
| 1584 |
+
"loss": 0.676,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 2.835633626097867,
|
| 1589 |
+
"grad_norm": 9.378284454345703,
|
| 1590 |
+
"learning_rate": 1.1041405269761608e-06,
|
| 1591 |
+
"loss": 0.6352,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 2.848180677540778,
|
| 1596 |
+
"grad_norm": 21.204431533813477,
|
| 1597 |
+
"learning_rate": 1.0204935173567545e-06,
|
| 1598 |
+
"loss": 0.7627,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 2.8607277289836888,
|
| 1603 |
+
"grad_norm": 23.616268157958984,
|
| 1604 |
+
"learning_rate": 9.368465077373484e-07,
|
| 1605 |
+
"loss": 0.6356,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 2.8732747804265997,
|
| 1610 |
+
"grad_norm": 25.798891067504883,
|
| 1611 |
+
"learning_rate": 8.531994981179424e-07,
|
| 1612 |
+
"loss": 0.7385,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 2.8858218318695106,
|
| 1617 |
+
"grad_norm": 11.757165908813477,
|
| 1618 |
+
"learning_rate": 7.695524884985363e-07,
|
| 1619 |
+
"loss": 0.6858,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 2.8983688833124215,
|
| 1624 |
+
"grad_norm": 9.811064720153809,
|
| 1625 |
+
"learning_rate": 6.859054788791301e-07,
|
| 1626 |
+
"loss": 0.6791,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 2.9109159347553324,
|
| 1631 |
+
"grad_norm": 14.021916389465332,
|
| 1632 |
+
"learning_rate": 6.02258469259724e-07,
|
| 1633 |
+
"loss": 0.8106,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 2.9234629861982433,
|
| 1638 |
+
"grad_norm": 16.58710479736328,
|
| 1639 |
+
"learning_rate": 5.186114596403179e-07,
|
| 1640 |
+
"loss": 0.8049,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 2.936010037641154,
|
| 1645 |
+
"grad_norm": 8.84688663482666,
|
| 1646 |
+
"learning_rate": 4.349644500209118e-07,
|
| 1647 |
+
"loss": 0.6731,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 2.948557089084065,
|
| 1652 |
+
"grad_norm": 9.54117488861084,
|
| 1653 |
+
"learning_rate": 3.5131744040150566e-07,
|
| 1654 |
+
"loss": 0.7293,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 2.961104140526976,
|
| 1659 |
+
"grad_norm": 17.779142379760742,
|
| 1660 |
+
"learning_rate": 2.6767043078209955e-07,
|
| 1661 |
+
"loss": 0.7803,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 2.973651191969887,
|
| 1666 |
+
"grad_norm": 7.330904006958008,
|
| 1667 |
+
"learning_rate": 1.8402342116269345e-07,
|
| 1668 |
+
"loss": 0.7281,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 2.9861982434127983,
|
| 1673 |
+
"grad_norm": 31.703744888305664,
|
| 1674 |
+
"learning_rate": 1.0037641154328732e-07,
|
| 1675 |
+
"loss": 0.8839,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 2.998745294855709,
|
| 1680 |
+
"grad_norm": 13.579190254211426,
|
| 1681 |
+
"learning_rate": 1.672940192388122e-08,
|
| 1682 |
+
"loss": 0.8382,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
}
|
| 1685 |
+
],
|
| 1686 |
+
"logging_steps": 10,
|
| 1687 |
+
"max_steps": 2391,
|
| 1688 |
+
"num_input_tokens_seen": 0,
|
| 1689 |
+
"num_train_epochs": 3,
|
| 1690 |
+
"save_steps": 500,
|
| 1691 |
+
"stateful_callbacks": {
|
| 1692 |
+
"TrainerControl": {
|
| 1693 |
+
"args": {
|
| 1694 |
+
"should_epoch_stop": false,
|
| 1695 |
+
"should_evaluate": false,
|
| 1696 |
+
"should_log": false,
|
| 1697 |
+
"should_save": true,
|
| 1698 |
+
"should_training_stop": true
|
| 1699 |
+
},
|
| 1700 |
+
"attributes": {}
|
| 1701 |
+
}
|
| 1702 |
+
},
|
| 1703 |
+
"total_flos": 274122632540160.0,
|
| 1704 |
+
"train_batch_size": 8,
|
| 1705 |
+
"trial_name": null,
|
| 1706 |
+
"trial_params": null
|
| 1707 |
+
}
|
checkpoint-2391/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ef9766f0d3350d702ce92717e829adb7fa49514a931cfa162fab4db7a275548d
|
| 3 |
+
size 5240
|