Training in progress, step 5000, checkpoint
Browse files
last-checkpoint/config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"UtteranceEmbedings"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"hidden_dropout_prob": 0.1,
|
| 7 |
+
"hidden_size": 768,
|
| 8 |
+
"intermediate_size": 3072,
|
| 9 |
+
"max_edu_length": 128,
|
| 10 |
+
"max_edus_per_dialog": 100,
|
| 11 |
+
"max_position_embeddings": 512,
|
| 12 |
+
"max_speakers": 200,
|
| 13 |
+
"model_type": "saute",
|
| 14 |
+
"num_attention_heads": 1,
|
| 15 |
+
"num_edu_layers": 2,
|
| 16 |
+
"num_hidden_layers": 1,
|
| 17 |
+
"num_speaker_embeddings": 512,
|
| 18 |
+
"num_token_layers": 2,
|
| 19 |
+
"speaker_embeddings_size": 768,
|
| 20 |
+
"torch_dtype": "float32",
|
| 21 |
+
"transformers_version": "4.52.4",
|
| 22 |
+
"vocab_size": 30522
|
| 23 |
+
}
|
last-checkpoint/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5055764d8bf73f79ca7a428dd271da991268e85a2d9756a29b72f1bbefb9104a
|
| 3 |
+
size 560983656
|
last-checkpoint/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f66e663ef1782e6f78f2a436a2d29f26ecfb1c44e1adbfb0b37aba90dd2d164
|
| 3 |
+
size 246053739
|
last-checkpoint/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:97ee020279a22b6910a66d68f6a36548720e0f741299e197d68a0300097035d6
|
| 3 |
+
size 14244
|
last-checkpoint/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0153ef62604f0a19d6db352c0efdd59e737e890fcf932261c593f4cb5173e799
|
| 3 |
+
size 988
|
last-checkpoint/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2715b803771bb9f0f36fe0c01e062993d36b56f916510bbea310e555c567b4cf
|
| 3 |
+
size 1064
|
last-checkpoint/trainer_state.json
ADDED
|
@@ -0,0 +1,1634 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.008392204648945687,
|
| 6 |
+
"eval_steps": 50,
|
| 7 |
+
"global_step": 5000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 8.392204648945687e-05,
|
| 14 |
+
"grad_norm": 3.729597806930542,
|
| 15 |
+
"learning_rate": 4.999588781972202e-05,
|
| 16 |
+
"loss": 1.6987,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 8.392204648945687e-05,
|
| 21 |
+
"eval_loss": 1.8344682455062866,
|
| 22 |
+
"eval_masked_accuracy": 0.6726457476615906,
|
| 23 |
+
"eval_runtime": 1.754,
|
| 24 |
+
"eval_samples_per_second": 5.701,
|
| 25 |
+
"eval_steps_per_second": 2.281,
|
| 26 |
+
"step": 50
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.00016784409297891374,
|
| 30 |
+
"grad_norm": 6.472073078155518,
|
| 31 |
+
"learning_rate": 4.999169171739755e-05,
|
| 32 |
+
"loss": 1.7415,
|
| 33 |
+
"step": 100
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.00016784409297891374,
|
| 37 |
+
"eval_loss": 1.7104336023330688,
|
| 38 |
+
"eval_masked_accuracy": 0.6737288236618042,
|
| 39 |
+
"eval_runtime": 1.7376,
|
| 40 |
+
"eval_samples_per_second": 5.755,
|
| 41 |
+
"eval_steps_per_second": 2.302,
|
| 42 |
+
"step": 100
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.0002517661394683706,
|
| 46 |
+
"grad_norm": 5.799453258514404,
|
| 47 |
+
"learning_rate": 4.998749561507307e-05,
|
| 48 |
+
"loss": 1.736,
|
| 49 |
+
"step": 150
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.0002517661394683706,
|
| 53 |
+
"eval_loss": 1.877158761024475,
|
| 54 |
+
"eval_masked_accuracy": 0.6936936974525452,
|
| 55 |
+
"eval_runtime": 1.7454,
|
| 56 |
+
"eval_samples_per_second": 5.729,
|
| 57 |
+
"eval_steps_per_second": 2.292,
|
| 58 |
+
"step": 150
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0003356881859578275,
|
| 62 |
+
"grad_norm": 9.896933555603027,
|
| 63 |
+
"learning_rate": 4.99832995127486e-05,
|
| 64 |
+
"loss": 1.7919,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.0003356881859578275,
|
| 69 |
+
"eval_loss": 1.424491047859192,
|
| 70 |
+
"eval_masked_accuracy": 0.7206477522850037,
|
| 71 |
+
"eval_runtime": 1.7315,
|
| 72 |
+
"eval_samples_per_second": 5.775,
|
| 73 |
+
"eval_steps_per_second": 2.31,
|
| 74 |
+
"step": 200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.00041961023244728435,
|
| 78 |
+
"grad_norm": 4.745198726654053,
|
| 79 |
+
"learning_rate": 4.997910341042413e-05,
|
| 80 |
+
"loss": 1.7252,
|
| 81 |
+
"step": 250
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.00041961023244728435,
|
| 85 |
+
"eval_loss": 1.915906310081482,
|
| 86 |
+
"eval_masked_accuracy": 0.6486486196517944,
|
| 87 |
+
"eval_runtime": 1.7431,
|
| 88 |
+
"eval_samples_per_second": 5.737,
|
| 89 |
+
"eval_steps_per_second": 2.295,
|
| 90 |
+
"step": 250
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.0005035322789367412,
|
| 94 |
+
"grad_norm": 6.004683971405029,
|
| 95 |
+
"learning_rate": 4.9974907308099657e-05,
|
| 96 |
+
"loss": 1.7487,
|
| 97 |
+
"step": 300
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.0005035322789367412,
|
| 101 |
+
"eval_loss": 1.7426478862762451,
|
| 102 |
+
"eval_masked_accuracy": 0.6846473217010498,
|
| 103 |
+
"eval_runtime": 1.7474,
|
| 104 |
+
"eval_samples_per_second": 5.723,
|
| 105 |
+
"eval_steps_per_second": 2.289,
|
| 106 |
+
"step": 300
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.0005874543254261981,
|
| 110 |
+
"grad_norm": 8.232338905334473,
|
| 111 |
+
"learning_rate": 4.9970711205775185e-05,
|
| 112 |
+
"loss": 1.6958,
|
| 113 |
+
"step": 350
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.0005874543254261981,
|
| 117 |
+
"eval_loss": 1.8806991577148438,
|
| 118 |
+
"eval_masked_accuracy": 0.6256157755851746,
|
| 119 |
+
"eval_runtime": 1.7452,
|
| 120 |
+
"eval_samples_per_second": 5.73,
|
| 121 |
+
"eval_steps_per_second": 2.292,
|
| 122 |
+
"step": 350
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.000671376371915655,
|
| 126 |
+
"grad_norm": 8.929485321044922,
|
| 127 |
+
"learning_rate": 4.996651510345071e-05,
|
| 128 |
+
"loss": 1.7165,
|
| 129 |
+
"step": 400
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.000671376371915655,
|
| 133 |
+
"eval_loss": 1.6669635772705078,
|
| 134 |
+
"eval_masked_accuracy": 0.6816326379776001,
|
| 135 |
+
"eval_runtime": 1.7367,
|
| 136 |
+
"eval_samples_per_second": 5.758,
|
| 137 |
+
"eval_steps_per_second": 2.303,
|
| 138 |
+
"step": 400
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.0007552984184051118,
|
| 142 |
+
"grad_norm": 6.171640872955322,
|
| 143 |
+
"learning_rate": 4.9962402923172725e-05,
|
| 144 |
+
"loss": 1.6222,
|
| 145 |
+
"step": 450
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.0007552984184051118,
|
| 149 |
+
"eval_loss": 2.174530506134033,
|
| 150 |
+
"eval_masked_accuracy": 0.6891891956329346,
|
| 151 |
+
"eval_runtime": 1.7554,
|
| 152 |
+
"eval_samples_per_second": 5.697,
|
| 153 |
+
"eval_steps_per_second": 2.279,
|
| 154 |
+
"step": 450
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.0008392204648945687,
|
| 158 |
+
"grad_norm": 4.092519283294678,
|
| 159 |
+
"learning_rate": 4.9958206820848254e-05,
|
| 160 |
+
"loss": 1.6441,
|
| 161 |
+
"step": 500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.0008392204648945687,
|
| 165 |
+
"eval_loss": 2.060279369354248,
|
| 166 |
+
"eval_masked_accuracy": 0.6461538672447205,
|
| 167 |
+
"eval_runtime": 1.8075,
|
| 168 |
+
"eval_samples_per_second": 5.532,
|
| 169 |
+
"eval_steps_per_second": 2.213,
|
| 170 |
+
"step": 500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.0009231425113840256,
|
| 174 |
+
"grad_norm": 5.34571647644043,
|
| 175 |
+
"learning_rate": 4.995401071852378e-05,
|
| 176 |
+
"loss": 1.7198,
|
| 177 |
+
"step": 550
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.0009231425113840256,
|
| 181 |
+
"eval_loss": 1.6280667781829834,
|
| 182 |
+
"eval_masked_accuracy": 0.6775510311126709,
|
| 183 |
+
"eval_runtime": 1.7487,
|
| 184 |
+
"eval_samples_per_second": 5.718,
|
| 185 |
+
"eval_steps_per_second": 2.287,
|
| 186 |
+
"step": 550
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.0010070645578734824,
|
| 190 |
+
"grad_norm": 4.286564350128174,
|
| 191 |
+
"learning_rate": 4.994981461619931e-05,
|
| 192 |
+
"loss": 1.6823,
|
| 193 |
+
"step": 600
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.0010070645578734824,
|
| 197 |
+
"eval_loss": 1.5270774364471436,
|
| 198 |
+
"eval_masked_accuracy": 0.6832579374313354,
|
| 199 |
+
"eval_runtime": 1.7459,
|
| 200 |
+
"eval_samples_per_second": 5.728,
|
| 201 |
+
"eval_steps_per_second": 2.291,
|
| 202 |
+
"step": 600
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.0010909866043629394,
|
| 206 |
+
"grad_norm": 3.7731900215148926,
|
| 207 |
+
"learning_rate": 4.994561851387484e-05,
|
| 208 |
+
"loss": 1.573,
|
| 209 |
+
"step": 650
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.0010909866043629394,
|
| 213 |
+
"eval_loss": 1.522475242614746,
|
| 214 |
+
"eval_masked_accuracy": 0.7423076629638672,
|
| 215 |
+
"eval_runtime": 1.7483,
|
| 216 |
+
"eval_samples_per_second": 5.72,
|
| 217 |
+
"eval_steps_per_second": 2.288,
|
| 218 |
+
"step": 650
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.0011749086508523962,
|
| 222 |
+
"grad_norm": 4.305816650390625,
|
| 223 |
+
"learning_rate": 4.994142241155036e-05,
|
| 224 |
+
"loss": 1.6905,
|
| 225 |
+
"step": 700
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.0011749086508523962,
|
| 229 |
+
"eval_loss": 1.503122091293335,
|
| 230 |
+
"eval_masked_accuracy": 0.67136150598526,
|
| 231 |
+
"eval_runtime": 1.7423,
|
| 232 |
+
"eval_samples_per_second": 5.74,
|
| 233 |
+
"eval_steps_per_second": 2.296,
|
| 234 |
+
"step": 700
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.0012588306973418532,
|
| 238 |
+
"grad_norm": 6.982117176055908,
|
| 239 |
+
"learning_rate": 4.993722630922589e-05,
|
| 240 |
+
"loss": 1.6444,
|
| 241 |
+
"step": 750
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.0012588306973418532,
|
| 245 |
+
"eval_loss": 1.7397890090942383,
|
| 246 |
+
"eval_masked_accuracy": 0.6946902871131897,
|
| 247 |
+
"eval_runtime": 1.7436,
|
| 248 |
+
"eval_samples_per_second": 5.735,
|
| 249 |
+
"eval_steps_per_second": 2.294,
|
| 250 |
+
"step": 750
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.00134275274383131,
|
| 254 |
+
"grad_norm": 6.332937717437744,
|
| 255 |
+
"learning_rate": 4.993303020690142e-05,
|
| 256 |
+
"loss": 1.7488,
|
| 257 |
+
"step": 800
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.00134275274383131,
|
| 261 |
+
"eval_loss": 1.407382845878601,
|
| 262 |
+
"eval_masked_accuracy": 0.7051281929016113,
|
| 263 |
+
"eval_runtime": 1.7398,
|
| 264 |
+
"eval_samples_per_second": 5.748,
|
| 265 |
+
"eval_steps_per_second": 2.299,
|
| 266 |
+
"step": 800
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.001426674790320767,
|
| 270 |
+
"grad_norm": 5.491461753845215,
|
| 271 |
+
"learning_rate": 4.9928834104576946e-05,
|
| 272 |
+
"loss": 1.5959,
|
| 273 |
+
"step": 850
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.001426674790320767,
|
| 277 |
+
"eval_loss": 1.8142907619476318,
|
| 278 |
+
"eval_masked_accuracy": 0.6625514626502991,
|
| 279 |
+
"eval_runtime": 1.7407,
|
| 280 |
+
"eval_samples_per_second": 5.745,
|
| 281 |
+
"eval_steps_per_second": 2.298,
|
| 282 |
+
"step": 850
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.0015105968368102237,
|
| 286 |
+
"grad_norm": 12.12775707244873,
|
| 287 |
+
"learning_rate": 4.9924638002252474e-05,
|
| 288 |
+
"loss": 1.6085,
|
| 289 |
+
"step": 900
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.0015105968368102237,
|
| 293 |
+
"eval_loss": 1.9904667139053345,
|
| 294 |
+
"eval_masked_accuracy": 0.6278026700019836,
|
| 295 |
+
"eval_runtime": 1.7503,
|
| 296 |
+
"eval_samples_per_second": 5.713,
|
| 297 |
+
"eval_steps_per_second": 2.285,
|
| 298 |
+
"step": 900
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.0015945188832996806,
|
| 302 |
+
"grad_norm": 18.452600479125977,
|
| 303 |
+
"learning_rate": 4.9920441899928e-05,
|
| 304 |
+
"loss": 1.5793,
|
| 305 |
+
"step": 950
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 0.0015945188832996806,
|
| 309 |
+
"eval_loss": 1.797326683998108,
|
| 310 |
+
"eval_masked_accuracy": 0.6784313917160034,
|
| 311 |
+
"eval_runtime": 1.7403,
|
| 312 |
+
"eval_samples_per_second": 5.746,
|
| 313 |
+
"eval_steps_per_second": 2.299,
|
| 314 |
+
"step": 950
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.0016784409297891374,
|
| 318 |
+
"grad_norm": 8.000075340270996,
|
| 319 |
+
"learning_rate": 4.9916245797603524e-05,
|
| 320 |
+
"loss": 1.5353,
|
| 321 |
+
"step": 1000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.0016784409297891374,
|
| 325 |
+
"eval_loss": 1.8558744192123413,
|
| 326 |
+
"eval_masked_accuracy": 0.6530612111091614,
|
| 327 |
+
"eval_runtime": 1.7574,
|
| 328 |
+
"eval_samples_per_second": 5.69,
|
| 329 |
+
"eval_steps_per_second": 2.276,
|
| 330 |
+
"step": 1000
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.0017623629762785944,
|
| 334 |
+
"grad_norm": 3.907064199447632,
|
| 335 |
+
"learning_rate": 4.991204969527905e-05,
|
| 336 |
+
"loss": 1.5363,
|
| 337 |
+
"step": 1050
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 0.0017623629762785944,
|
| 341 |
+
"eval_loss": 2.0765745639801025,
|
| 342 |
+
"eval_masked_accuracy": 0.6553191542625427,
|
| 343 |
+
"eval_runtime": 1.7983,
|
| 344 |
+
"eval_samples_per_second": 5.561,
|
| 345 |
+
"eval_steps_per_second": 2.224,
|
| 346 |
+
"step": 1050
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.0018462850227680511,
|
| 350 |
+
"grad_norm": 4.185476303100586,
|
| 351 |
+
"learning_rate": 4.990785359295458e-05,
|
| 352 |
+
"loss": 1.6641,
|
| 353 |
+
"step": 1100
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.0018462850227680511,
|
| 357 |
+
"eval_loss": 1.5849405527114868,
|
| 358 |
+
"eval_masked_accuracy": 0.71074378490448,
|
| 359 |
+
"eval_runtime": 1.7601,
|
| 360 |
+
"eval_samples_per_second": 5.681,
|
| 361 |
+
"eval_steps_per_second": 2.273,
|
| 362 |
+
"step": 1100
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.0019302070692575081,
|
| 366 |
+
"grad_norm": 5.447309494018555,
|
| 367 |
+
"learning_rate": 4.990365749063011e-05,
|
| 368 |
+
"loss": 1.7069,
|
| 369 |
+
"step": 1150
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.0019302070692575081,
|
| 373 |
+
"eval_loss": 1.6813358068466187,
|
| 374 |
+
"eval_masked_accuracy": 0.7231404781341553,
|
| 375 |
+
"eval_runtime": 1.7529,
|
| 376 |
+
"eval_samples_per_second": 5.705,
|
| 377 |
+
"eval_steps_per_second": 2.282,
|
| 378 |
+
"step": 1150
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 0.002014129115746965,
|
| 382 |
+
"grad_norm": 5.904290199279785,
|
| 383 |
+
"learning_rate": 4.989946138830564e-05,
|
| 384 |
+
"loss": 1.6996,
|
| 385 |
+
"step": 1200
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 0.002014129115746965,
|
| 389 |
+
"eval_loss": 1.6986854076385498,
|
| 390 |
+
"eval_masked_accuracy": 0.6554622054100037,
|
| 391 |
+
"eval_runtime": 1.7531,
|
| 392 |
+
"eval_samples_per_second": 5.704,
|
| 393 |
+
"eval_steps_per_second": 2.282,
|
| 394 |
+
"step": 1200
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.002098051162236422,
|
| 398 |
+
"grad_norm": 5.6478986740112305,
|
| 399 |
+
"learning_rate": 4.989526528598116e-05,
|
| 400 |
+
"loss": 1.5291,
|
| 401 |
+
"step": 1250
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.002098051162236422,
|
| 405 |
+
"eval_loss": 1.7059627771377563,
|
| 406 |
+
"eval_masked_accuracy": 0.6680498123168945,
|
| 407 |
+
"eval_runtime": 1.7416,
|
| 408 |
+
"eval_samples_per_second": 5.742,
|
| 409 |
+
"eval_steps_per_second": 2.297,
|
| 410 |
+
"step": 1250
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.002181973208725879,
|
| 414 |
+
"grad_norm": 6.695890426635742,
|
| 415 |
+
"learning_rate": 4.989106918365669e-05,
|
| 416 |
+
"loss": 1.8386,
|
| 417 |
+
"step": 1300
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.002181973208725879,
|
| 421 |
+
"eval_loss": 1.6500450372695923,
|
| 422 |
+
"eval_masked_accuracy": 0.6693877577781677,
|
| 423 |
+
"eval_runtime": 1.7414,
|
| 424 |
+
"eval_samples_per_second": 5.743,
|
| 425 |
+
"eval_steps_per_second": 2.297,
|
| 426 |
+
"step": 1300
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.0022658952552153354,
|
| 430 |
+
"grad_norm": 4.831510066986084,
|
| 431 |
+
"learning_rate": 4.9886873081332217e-05,
|
| 432 |
+
"loss": 1.691,
|
| 433 |
+
"step": 1350
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.0022658952552153354,
|
| 437 |
+
"eval_loss": 1.4610856771469116,
|
| 438 |
+
"eval_masked_accuracy": 0.7090163826942444,
|
| 439 |
+
"eval_runtime": 1.7413,
|
| 440 |
+
"eval_samples_per_second": 5.743,
|
| 441 |
+
"eval_steps_per_second": 2.297,
|
| 442 |
+
"step": 1350
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.0023498173017047924,
|
| 446 |
+
"grad_norm": 4.90496826171875,
|
| 447 |
+
"learning_rate": 4.9882676979007745e-05,
|
| 448 |
+
"loss": 1.7116,
|
| 449 |
+
"step": 1400
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.0023498173017047924,
|
| 453 |
+
"eval_loss": 1.6787996292114258,
|
| 454 |
+
"eval_masked_accuracy": 0.6153846383094788,
|
| 455 |
+
"eval_runtime": 1.7655,
|
| 456 |
+
"eval_samples_per_second": 5.664,
|
| 457 |
+
"eval_steps_per_second": 2.266,
|
| 458 |
+
"step": 1400
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.0024337393481942493,
|
| 462 |
+
"grad_norm": 5.956592559814453,
|
| 463 |
+
"learning_rate": 4.9878480876683273e-05,
|
| 464 |
+
"loss": 1.5348,
|
| 465 |
+
"step": 1450
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.0024337393481942493,
|
| 469 |
+
"eval_loss": 1.7995752096176147,
|
| 470 |
+
"eval_masked_accuracy": 0.6759999990463257,
|
| 471 |
+
"eval_runtime": 1.7486,
|
| 472 |
+
"eval_samples_per_second": 5.719,
|
| 473 |
+
"eval_steps_per_second": 2.288,
|
| 474 |
+
"step": 1450
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.0025176613946837063,
|
| 478 |
+
"grad_norm": 5.731600761413574,
|
| 479 |
+
"learning_rate": 4.9874284774358795e-05,
|
| 480 |
+
"loss": 1.5617,
|
| 481 |
+
"step": 1500
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 0.0025176613946837063,
|
| 485 |
+
"eval_loss": 2.028412342071533,
|
| 486 |
+
"eval_masked_accuracy": 0.6007905006408691,
|
| 487 |
+
"eval_runtime": 1.797,
|
| 488 |
+
"eval_samples_per_second": 5.565,
|
| 489 |
+
"eval_steps_per_second": 2.226,
|
| 490 |
+
"step": 1500
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.002601583441173163,
|
| 494 |
+
"grad_norm": 9.261569023132324,
|
| 495 |
+
"learning_rate": 4.9870088672034324e-05,
|
| 496 |
+
"loss": 1.7109,
|
| 497 |
+
"step": 1550
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.002601583441173163,
|
| 501 |
+
"eval_loss": 1.8843729496002197,
|
| 502 |
+
"eval_masked_accuracy": 0.6594203114509583,
|
| 503 |
+
"eval_runtime": 1.7575,
|
| 504 |
+
"eval_samples_per_second": 5.69,
|
| 505 |
+
"eval_steps_per_second": 2.276,
|
| 506 |
+
"step": 1550
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.00268550548766262,
|
| 510 |
+
"grad_norm": 7.181281089782715,
|
| 511 |
+
"learning_rate": 4.986589256970985e-05,
|
| 512 |
+
"loss": 1.6529,
|
| 513 |
+
"step": 1600
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.00268550548766262,
|
| 517 |
+
"eval_loss": 1.5639550685882568,
|
| 518 |
+
"eval_masked_accuracy": 0.6905829310417175,
|
| 519 |
+
"eval_runtime": 1.7429,
|
| 520 |
+
"eval_samples_per_second": 5.738,
|
| 521 |
+
"eval_steps_per_second": 2.295,
|
| 522 |
+
"step": 1600
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.002769427534152077,
|
| 526 |
+
"grad_norm": 5.245086193084717,
|
| 527 |
+
"learning_rate": 4.986169646738538e-05,
|
| 528 |
+
"loss": 1.6497,
|
| 529 |
+
"step": 1650
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.002769427534152077,
|
| 533 |
+
"eval_loss": 1.4776060581207275,
|
| 534 |
+
"eval_masked_accuracy": 0.7312775254249573,
|
| 535 |
+
"eval_runtime": 1.7496,
|
| 536 |
+
"eval_samples_per_second": 5.715,
|
| 537 |
+
"eval_steps_per_second": 2.286,
|
| 538 |
+
"step": 1650
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 0.002853349580641534,
|
| 542 |
+
"grad_norm": 5.593554496765137,
|
| 543 |
+
"learning_rate": 4.98575003650609e-05,
|
| 544 |
+
"loss": 1.5326,
|
| 545 |
+
"step": 1700
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 0.002853349580641534,
|
| 549 |
+
"eval_loss": 2.0159146785736084,
|
| 550 |
+
"eval_masked_accuracy": 0.6205357313156128,
|
| 551 |
+
"eval_runtime": 1.7381,
|
| 552 |
+
"eval_samples_per_second": 5.753,
|
| 553 |
+
"eval_steps_per_second": 2.301,
|
| 554 |
+
"step": 1700
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.0029372716271309908,
|
| 558 |
+
"grad_norm": 7.406851768493652,
|
| 559 |
+
"learning_rate": 4.985330426273643e-05,
|
| 560 |
+
"loss": 1.5081,
|
| 561 |
+
"step": 1750
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.0029372716271309908,
|
| 565 |
+
"eval_loss": 1.508250117301941,
|
| 566 |
+
"eval_masked_accuracy": 0.7027027010917664,
|
| 567 |
+
"eval_runtime": 1.7474,
|
| 568 |
+
"eval_samples_per_second": 5.723,
|
| 569 |
+
"eval_steps_per_second": 2.289,
|
| 570 |
+
"step": 1750
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.0030211936736204473,
|
| 574 |
+
"grad_norm": 4.5133514404296875,
|
| 575 |
+
"learning_rate": 4.984910816041196e-05,
|
| 576 |
+
"loss": 1.6619,
|
| 577 |
+
"step": 1800
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.0030211936736204473,
|
| 581 |
+
"eval_loss": 1.7022559642791748,
|
| 582 |
+
"eval_masked_accuracy": 0.694779098033905,
|
| 583 |
+
"eval_runtime": 1.748,
|
| 584 |
+
"eval_samples_per_second": 5.721,
|
| 585 |
+
"eval_steps_per_second": 2.288,
|
| 586 |
+
"step": 1800
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.0031051157201099043,
|
| 590 |
+
"grad_norm": 7.173299312591553,
|
| 591 |
+
"learning_rate": 4.984491205808749e-05,
|
| 592 |
+
"loss": 1.7603,
|
| 593 |
+
"step": 1850
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 0.0031051157201099043,
|
| 597 |
+
"eval_loss": 1.6458946466445923,
|
| 598 |
+
"eval_masked_accuracy": 0.6636771559715271,
|
| 599 |
+
"eval_runtime": 1.8607,
|
| 600 |
+
"eval_samples_per_second": 5.374,
|
| 601 |
+
"eval_steps_per_second": 2.15,
|
| 602 |
+
"step": 1850
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.0031890377665993613,
|
| 606 |
+
"grad_norm": 4.3678975105285645,
|
| 607 |
+
"learning_rate": 4.984071595576301e-05,
|
| 608 |
+
"loss": 1.6453,
|
| 609 |
+
"step": 1900
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.0031890377665993613,
|
| 613 |
+
"eval_loss": 1.8176072835922241,
|
| 614 |
+
"eval_masked_accuracy": 0.6724137663841248,
|
| 615 |
+
"eval_runtime": 1.7589,
|
| 616 |
+
"eval_samples_per_second": 5.685,
|
| 617 |
+
"eval_steps_per_second": 2.274,
|
| 618 |
+
"step": 1900
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.0032729598130888183,
|
| 622 |
+
"grad_norm": 7.378585338592529,
|
| 623 |
+
"learning_rate": 4.983651985343854e-05,
|
| 624 |
+
"loss": 1.6409,
|
| 625 |
+
"step": 1950
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.0032729598130888183,
|
| 629 |
+
"eval_loss": 2.0491786003112793,
|
| 630 |
+
"eval_masked_accuracy": 0.6374502182006836,
|
| 631 |
+
"eval_runtime": 1.757,
|
| 632 |
+
"eval_samples_per_second": 5.692,
|
| 633 |
+
"eval_steps_per_second": 2.277,
|
| 634 |
+
"step": 1950
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.003356881859578275,
|
| 638 |
+
"grad_norm": 4.898635387420654,
|
| 639 |
+
"learning_rate": 4.9832323751114066e-05,
|
| 640 |
+
"loss": 1.6994,
|
| 641 |
+
"step": 2000
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.003356881859578275,
|
| 645 |
+
"eval_loss": 1.4773211479187012,
|
| 646 |
+
"eval_masked_accuracy": 0.6739130616188049,
|
| 647 |
+
"eval_runtime": 1.7439,
|
| 648 |
+
"eval_samples_per_second": 5.734,
|
| 649 |
+
"eval_steps_per_second": 2.294,
|
| 650 |
+
"step": 2000
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.003440803906067732,
|
| 654 |
+
"grad_norm": 7.465532302856445,
|
| 655 |
+
"learning_rate": 4.9828127648789594e-05,
|
| 656 |
+
"loss": 1.5798,
|
| 657 |
+
"step": 2050
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.003440803906067732,
|
| 661 |
+
"eval_loss": 1.6743123531341553,
|
| 662 |
+
"eval_masked_accuracy": 0.6770427823066711,
|
| 663 |
+
"eval_runtime": 1.7546,
|
| 664 |
+
"eval_samples_per_second": 5.699,
|
| 665 |
+
"eval_steps_per_second": 2.28,
|
| 666 |
+
"step": 2050
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 0.0035247259525571888,
|
| 670 |
+
"grad_norm": 7.025172233581543,
|
| 671 |
+
"learning_rate": 4.982393154646512e-05,
|
| 672 |
+
"loss": 1.7312,
|
| 673 |
+
"step": 2100
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 0.0035247259525571888,
|
| 677 |
+
"eval_loss": 1.726737380027771,
|
| 678 |
+
"eval_masked_accuracy": 0.6824034452438354,
|
| 679 |
+
"eval_runtime": 1.7272,
|
| 680 |
+
"eval_samples_per_second": 5.79,
|
| 681 |
+
"eval_steps_per_second": 2.316,
|
| 682 |
+
"step": 2100
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.0036086479990466457,
|
| 686 |
+
"grad_norm": 8.405756950378418,
|
| 687 |
+
"learning_rate": 4.9819735444140644e-05,
|
| 688 |
+
"loss": 1.7284,
|
| 689 |
+
"step": 2150
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.0036086479990466457,
|
| 693 |
+
"eval_loss": 1.8043725490570068,
|
| 694 |
+
"eval_masked_accuracy": 0.6153846383094788,
|
| 695 |
+
"eval_runtime": 1.7569,
|
| 696 |
+
"eval_samples_per_second": 5.692,
|
| 697 |
+
"eval_steps_per_second": 2.277,
|
| 698 |
+
"step": 2150
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.0036925700455361023,
|
| 702 |
+
"grad_norm": 6.279454231262207,
|
| 703 |
+
"learning_rate": 4.981553934181617e-05,
|
| 704 |
+
"loss": 1.4629,
|
| 705 |
+
"step": 2200
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.0036925700455361023,
|
| 709 |
+
"eval_loss": 1.8529506921768188,
|
| 710 |
+
"eval_masked_accuracy": 0.6823529601097107,
|
| 711 |
+
"eval_runtime": 1.7798,
|
| 712 |
+
"eval_samples_per_second": 5.619,
|
| 713 |
+
"eval_steps_per_second": 2.247,
|
| 714 |
+
"step": 2200
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.0037764920920255593,
|
| 718 |
+
"grad_norm": 7.525041580200195,
|
| 719 |
+
"learning_rate": 4.98113432394917e-05,
|
| 720 |
+
"loss": 1.5309,
|
| 721 |
+
"step": 2250
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.0037764920920255593,
|
| 725 |
+
"eval_loss": 1.8144168853759766,
|
| 726 |
+
"eval_masked_accuracy": 0.7272727489471436,
|
| 727 |
+
"eval_runtime": 1.7418,
|
| 728 |
+
"eval_samples_per_second": 5.741,
|
| 729 |
+
"eval_steps_per_second": 2.297,
|
| 730 |
+
"step": 2250
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.0038604141385150162,
|
| 734 |
+
"grad_norm": 6.561546802520752,
|
| 735 |
+
"learning_rate": 4.980714713716723e-05,
|
| 736 |
+
"loss": 1.6761,
|
| 737 |
+
"step": 2300
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.0038604141385150162,
|
| 741 |
+
"eval_loss": 1.8419244289398193,
|
| 742 |
+
"eval_masked_accuracy": 0.6638298034667969,
|
| 743 |
+
"eval_runtime": 1.7921,
|
| 744 |
+
"eval_samples_per_second": 5.58,
|
| 745 |
+
"eval_steps_per_second": 2.232,
|
| 746 |
+
"step": 2300
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.003944336185004473,
|
| 750 |
+
"grad_norm": 4.7332987785339355,
|
| 751 |
+
"learning_rate": 4.980295103484276e-05,
|
| 752 |
+
"loss": 1.6738,
|
| 753 |
+
"step": 2350
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.003944336185004473,
|
| 757 |
+
"eval_loss": 1.576249122619629,
|
| 758 |
+
"eval_masked_accuracy": 0.7078651785850525,
|
| 759 |
+
"eval_runtime": 1.7414,
|
| 760 |
+
"eval_samples_per_second": 5.742,
|
| 761 |
+
"eval_steps_per_second": 2.297,
|
| 762 |
+
"step": 2350
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 0.00402825823149393,
|
| 766 |
+
"grad_norm": 3.7719192504882812,
|
| 767 |
+
"learning_rate": 4.979875493251828e-05,
|
| 768 |
+
"loss": 1.6432,
|
| 769 |
+
"step": 2400
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 0.00402825823149393,
|
| 773 |
+
"eval_loss": 1.811785340309143,
|
| 774 |
+
"eval_masked_accuracy": 0.6746031641960144,
|
| 775 |
+
"eval_runtime": 1.7463,
|
| 776 |
+
"eval_samples_per_second": 5.726,
|
| 777 |
+
"eval_steps_per_second": 2.291,
|
| 778 |
+
"step": 2400
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.004112180277983387,
|
| 782 |
+
"grad_norm": 6.218490123748779,
|
| 783 |
+
"learning_rate": 4.979455883019381e-05,
|
| 784 |
+
"loss": 1.5416,
|
| 785 |
+
"step": 2450
|
| 786 |
+
},
|
| 787 |
+
{
|
| 788 |
+
"epoch": 0.004112180277983387,
|
| 789 |
+
"eval_loss": 1.6883758306503296,
|
| 790 |
+
"eval_masked_accuracy": 0.6900826692581177,
|
| 791 |
+
"eval_runtime": 1.7374,
|
| 792 |
+
"eval_samples_per_second": 5.756,
|
| 793 |
+
"eval_steps_per_second": 2.302,
|
| 794 |
+
"step": 2450
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.004196102324472844,
|
| 798 |
+
"grad_norm": 5.042550086975098,
|
| 799 |
+
"learning_rate": 4.9790362727869336e-05,
|
| 800 |
+
"loss": 1.6701,
|
| 801 |
+
"step": 2500
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.004196102324472844,
|
| 805 |
+
"eval_loss": 1.567375898361206,
|
| 806 |
+
"eval_masked_accuracy": 0.6788617968559265,
|
| 807 |
+
"eval_runtime": 1.735,
|
| 808 |
+
"eval_samples_per_second": 5.764,
|
| 809 |
+
"eval_steps_per_second": 2.306,
|
| 810 |
+
"step": 2500
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.0042800243709623,
|
| 814 |
+
"grad_norm": 6.665520668029785,
|
| 815 |
+
"learning_rate": 4.9786166625544865e-05,
|
| 816 |
+
"loss": 1.6006,
|
| 817 |
+
"step": 2550
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 0.0042800243709623,
|
| 821 |
+
"eval_loss": 1.659168004989624,
|
| 822 |
+
"eval_masked_accuracy": 0.6385542154312134,
|
| 823 |
+
"eval_runtime": 1.7434,
|
| 824 |
+
"eval_samples_per_second": 5.736,
|
| 825 |
+
"eval_steps_per_second": 2.294,
|
| 826 |
+
"step": 2550
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.004363946417451758,
|
| 830 |
+
"grad_norm": 4.378693580627441,
|
| 831 |
+
"learning_rate": 4.978197052322039e-05,
|
| 832 |
+
"loss": 1.6363,
|
| 833 |
+
"step": 2600
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 0.004363946417451758,
|
| 837 |
+
"eval_loss": 1.6367610692977905,
|
| 838 |
+
"eval_masked_accuracy": 0.6679389476776123,
|
| 839 |
+
"eval_runtime": 1.7407,
|
| 840 |
+
"eval_samples_per_second": 5.745,
|
| 841 |
+
"eval_steps_per_second": 2.298,
|
| 842 |
+
"step": 2600
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.004447868463941214,
|
| 846 |
+
"grad_norm": 8.087454795837402,
|
| 847 |
+
"learning_rate": 4.977777442089592e-05,
|
| 848 |
+
"loss": 1.5518,
|
| 849 |
+
"step": 2650
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.004447868463941214,
|
| 853 |
+
"eval_loss": 2.035369873046875,
|
| 854 |
+
"eval_masked_accuracy": 0.64462810754776,
|
| 855 |
+
"eval_runtime": 1.7443,
|
| 856 |
+
"eval_samples_per_second": 5.733,
|
| 857 |
+
"eval_steps_per_second": 2.293,
|
| 858 |
+
"step": 2650
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.004531790510430671,
|
| 862 |
+
"grad_norm": 6.383141040802002,
|
| 863 |
+
"learning_rate": 4.977357831857144e-05,
|
| 864 |
+
"loss": 1.7726,
|
| 865 |
+
"step": 2700
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.004531790510430671,
|
| 869 |
+
"eval_loss": 1.9139858484268188,
|
| 870 |
+
"eval_masked_accuracy": 0.7137096524238586,
|
| 871 |
+
"eval_runtime": 1.7376,
|
| 872 |
+
"eval_samples_per_second": 5.755,
|
| 873 |
+
"eval_steps_per_second": 2.302,
|
| 874 |
+
"step": 2700
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.004615712556920128,
|
| 878 |
+
"grad_norm": 8.098458290100098,
|
| 879 |
+
"learning_rate": 4.976938221624697e-05,
|
| 880 |
+
"loss": 1.701,
|
| 881 |
+
"step": 2750
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.004615712556920128,
|
| 885 |
+
"eval_loss": 1.8784687519073486,
|
| 886 |
+
"eval_masked_accuracy": 0.6525096297264099,
|
| 887 |
+
"eval_runtime": 1.8538,
|
| 888 |
+
"eval_samples_per_second": 5.394,
|
| 889 |
+
"eval_steps_per_second": 2.158,
|
| 890 |
+
"step": 2750
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.004699634603409585,
|
| 894 |
+
"grad_norm": 5.3736138343811035,
|
| 895 |
+
"learning_rate": 4.97651861139225e-05,
|
| 896 |
+
"loss": 1.5577,
|
| 897 |
+
"step": 2800
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.004699634603409585,
|
| 901 |
+
"eval_loss": 1.6022107601165771,
|
| 902 |
+
"eval_masked_accuracy": 0.6907630562782288,
|
| 903 |
+
"eval_runtime": 1.7368,
|
| 904 |
+
"eval_samples_per_second": 5.758,
|
| 905 |
+
"eval_steps_per_second": 2.303,
|
| 906 |
+
"step": 2800
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.004783556649899042,
|
| 910 |
+
"grad_norm": 4.617998123168945,
|
| 911 |
+
"learning_rate": 4.976099001159803e-05,
|
| 912 |
+
"loss": 1.6194,
|
| 913 |
+
"step": 2850
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.004783556649899042,
|
| 917 |
+
"eval_loss": 1.398147702217102,
|
| 918 |
+
"eval_masked_accuracy": 0.6696035265922546,
|
| 919 |
+
"eval_runtime": 1.7507,
|
| 920 |
+
"eval_samples_per_second": 5.712,
|
| 921 |
+
"eval_steps_per_second": 2.285,
|
| 922 |
+
"step": 2850
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.004867478696388499,
|
| 926 |
+
"grad_norm": 4.976247787475586,
|
| 927 |
+
"learning_rate": 4.975679390927356e-05,
|
| 928 |
+
"loss": 1.6325,
|
| 929 |
+
"step": 2900
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 0.004867478696388499,
|
| 933 |
+
"eval_loss": 1.7178815603256226,
|
| 934 |
+
"eval_masked_accuracy": 0.6653696298599243,
|
| 935 |
+
"eval_runtime": 1.7533,
|
| 936 |
+
"eval_samples_per_second": 5.703,
|
| 937 |
+
"eval_steps_per_second": 2.281,
|
| 938 |
+
"step": 2900
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 0.004951400742877955,
|
| 942 |
+
"grad_norm": 5.229081153869629,
|
| 943 |
+
"learning_rate": 4.9752597806949085e-05,
|
| 944 |
+
"loss": 1.7057,
|
| 945 |
+
"step": 2950
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 0.004951400742877955,
|
| 949 |
+
"eval_loss": 1.8161494731903076,
|
| 950 |
+
"eval_masked_accuracy": 0.6431535482406616,
|
| 951 |
+
"eval_runtime": 1.7382,
|
| 952 |
+
"eval_samples_per_second": 5.753,
|
| 953 |
+
"eval_steps_per_second": 2.301,
|
| 954 |
+
"step": 2950
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.005035322789367413,
|
| 958 |
+
"grad_norm": 6.112144947052002,
|
| 959 |
+
"learning_rate": 4.974840170462461e-05,
|
| 960 |
+
"loss": 1.6189,
|
| 961 |
+
"step": 3000
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.005035322789367413,
|
| 965 |
+
"eval_loss": 1.8454160690307617,
|
| 966 |
+
"eval_masked_accuracy": 0.6767241358757019,
|
| 967 |
+
"eval_runtime": 1.7376,
|
| 968 |
+
"eval_samples_per_second": 5.755,
|
| 969 |
+
"eval_steps_per_second": 2.302,
|
| 970 |
+
"step": 3000
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.005119244835856869,
|
| 974 |
+
"grad_norm": 5.7698445320129395,
|
| 975 |
+
"learning_rate": 4.9744205602300135e-05,
|
| 976 |
+
"loss": 1.6734,
|
| 977 |
+
"step": 3050
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 0.005119244835856869,
|
| 981 |
+
"eval_loss": 1.6155188083648682,
|
| 982 |
+
"eval_masked_accuracy": 0.6991525292396545,
|
| 983 |
+
"eval_runtime": 1.7469,
|
| 984 |
+
"eval_samples_per_second": 5.724,
|
| 985 |
+
"eval_steps_per_second": 2.29,
|
| 986 |
+
"step": 3050
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.005203166882346326,
|
| 990 |
+
"grad_norm": 11.4446382522583,
|
| 991 |
+
"learning_rate": 4.9740009499975664e-05,
|
| 992 |
+
"loss": 1.602,
|
| 993 |
+
"step": 3100
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 0.005203166882346326,
|
| 997 |
+
"eval_loss": 1.7193024158477783,
|
| 998 |
+
"eval_masked_accuracy": 0.6454545259475708,
|
| 999 |
+
"eval_runtime": 1.8085,
|
| 1000 |
+
"eval_samples_per_second": 5.529,
|
| 1001 |
+
"eval_steps_per_second": 2.212,
|
| 1002 |
+
"step": 3100
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.005287088928835783,
|
| 1006 |
+
"grad_norm": 4.331955432891846,
|
| 1007 |
+
"learning_rate": 4.973581339765119e-05,
|
| 1008 |
+
"loss": 1.5886,
|
| 1009 |
+
"step": 3150
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 0.005287088928835783,
|
| 1013 |
+
"eval_loss": 1.7239084243774414,
|
| 1014 |
+
"eval_masked_accuracy": 0.7025862336158752,
|
| 1015 |
+
"eval_runtime": 1.7507,
|
| 1016 |
+
"eval_samples_per_second": 5.712,
|
| 1017 |
+
"eval_steps_per_second": 2.285,
|
| 1018 |
+
"step": 3150
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.00537101097532524,
|
| 1022 |
+
"grad_norm": 6.857669830322266,
|
| 1023 |
+
"learning_rate": 4.973161729532672e-05,
|
| 1024 |
+
"loss": 1.6531,
|
| 1025 |
+
"step": 3200
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.00537101097532524,
|
| 1029 |
+
"eval_loss": 1.7898776531219482,
|
| 1030 |
+
"eval_masked_accuracy": 0.6463878154754639,
|
| 1031 |
+
"eval_runtime": 1.807,
|
| 1032 |
+
"eval_samples_per_second": 5.534,
|
| 1033 |
+
"eval_steps_per_second": 2.214,
|
| 1034 |
+
"step": 3200
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 0.005454933021814697,
|
| 1038 |
+
"grad_norm": 6.366724491119385,
|
| 1039 |
+
"learning_rate": 4.972742119300224e-05,
|
| 1040 |
+
"loss": 1.5112,
|
| 1041 |
+
"step": 3250
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 0.005454933021814697,
|
| 1045 |
+
"eval_loss": 1.68304443359375,
|
| 1046 |
+
"eval_masked_accuracy": 0.6958174705505371,
|
| 1047 |
+
"eval_runtime": 1.7544,
|
| 1048 |
+
"eval_samples_per_second": 5.7,
|
| 1049 |
+
"eval_steps_per_second": 2.28,
|
| 1050 |
+
"step": 3250
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 0.005538855068304154,
|
| 1054 |
+
"grad_norm": 5.657731056213379,
|
| 1055 |
+
"learning_rate": 4.972322509067777e-05,
|
| 1056 |
+
"loss": 1.5622,
|
| 1057 |
+
"step": 3300
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 0.005538855068304154,
|
| 1061 |
+
"eval_loss": 1.7854249477386475,
|
| 1062 |
+
"eval_masked_accuracy": 0.6833333373069763,
|
| 1063 |
+
"eval_runtime": 1.7977,
|
| 1064 |
+
"eval_samples_per_second": 5.563,
|
| 1065 |
+
"eval_steps_per_second": 2.225,
|
| 1066 |
+
"step": 3300
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.00562277711479361,
|
| 1070 |
+
"grad_norm": 4.501428127288818,
|
| 1071 |
+
"learning_rate": 4.97190289883533e-05,
|
| 1072 |
+
"loss": 1.5736,
|
| 1073 |
+
"step": 3350
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.00562277711479361,
|
| 1077 |
+
"eval_loss": 1.4276224374771118,
|
| 1078 |
+
"eval_masked_accuracy": 0.7192118167877197,
|
| 1079 |
+
"eval_runtime": 1.7643,
|
| 1080 |
+
"eval_samples_per_second": 5.668,
|
| 1081 |
+
"eval_steps_per_second": 2.267,
|
| 1082 |
+
"step": 3350
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 0.005706699161283068,
|
| 1086 |
+
"grad_norm": 6.436139106750488,
|
| 1087 |
+
"learning_rate": 4.971483288602883e-05,
|
| 1088 |
+
"loss": 1.5653,
|
| 1089 |
+
"step": 3400
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.005706699161283068,
|
| 1093 |
+
"eval_loss": 1.674355149269104,
|
| 1094 |
+
"eval_masked_accuracy": 0.718367338180542,
|
| 1095 |
+
"eval_runtime": 1.7482,
|
| 1096 |
+
"eval_samples_per_second": 5.72,
|
| 1097 |
+
"eval_steps_per_second": 2.288,
|
| 1098 |
+
"step": 3400
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 0.005790621207772524,
|
| 1102 |
+
"grad_norm": 6.295548439025879,
|
| 1103 |
+
"learning_rate": 4.9710636783704356e-05,
|
| 1104 |
+
"loss": 1.5556,
|
| 1105 |
+
"step": 3450
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 0.005790621207772524,
|
| 1109 |
+
"eval_loss": 1.7501426935195923,
|
| 1110 |
+
"eval_masked_accuracy": 0.7076271176338196,
|
| 1111 |
+
"eval_runtime": 1.7554,
|
| 1112 |
+
"eval_samples_per_second": 5.697,
|
| 1113 |
+
"eval_steps_per_second": 2.279,
|
| 1114 |
+
"step": 3450
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 0.0058745432542619816,
|
| 1118 |
+
"grad_norm": 5.733904838562012,
|
| 1119 |
+
"learning_rate": 4.9706440681379884e-05,
|
| 1120 |
+
"loss": 1.5164,
|
| 1121 |
+
"step": 3500
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 0.0058745432542619816,
|
| 1125 |
+
"eval_loss": 1.520179033279419,
|
| 1126 |
+
"eval_masked_accuracy": 0.7203390002250671,
|
| 1127 |
+
"eval_runtime": 1.7629,
|
| 1128 |
+
"eval_samples_per_second": 5.672,
|
| 1129 |
+
"eval_steps_per_second": 2.269,
|
| 1130 |
+
"step": 3500
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.005958465300751438,
|
| 1134 |
+
"grad_norm": 5.285616397857666,
|
| 1135 |
+
"learning_rate": 4.9702244579055406e-05,
|
| 1136 |
+
"loss": 1.6254,
|
| 1137 |
+
"step": 3550
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.005958465300751438,
|
| 1141 |
+
"eval_loss": 1.7321217060089111,
|
| 1142 |
+
"eval_masked_accuracy": 0.6712962985038757,
|
| 1143 |
+
"eval_runtime": 1.7429,
|
| 1144 |
+
"eval_samples_per_second": 5.738,
|
| 1145 |
+
"eval_steps_per_second": 2.295,
|
| 1146 |
+
"step": 3550
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 0.006042387347240895,
|
| 1150 |
+
"grad_norm": 5.386379241943359,
|
| 1151 |
+
"learning_rate": 4.9698048476730934e-05,
|
| 1152 |
+
"loss": 1.505,
|
| 1153 |
+
"step": 3600
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 0.006042387347240895,
|
| 1157 |
+
"eval_loss": 1.7810560464859009,
|
| 1158 |
+
"eval_masked_accuracy": 0.654618501663208,
|
| 1159 |
+
"eval_runtime": 1.7454,
|
| 1160 |
+
"eval_samples_per_second": 5.729,
|
| 1161 |
+
"eval_steps_per_second": 2.292,
|
| 1162 |
+
"step": 3600
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 0.006126309393730352,
|
| 1166 |
+
"grad_norm": 6.726806640625,
|
| 1167 |
+
"learning_rate": 4.969385237440646e-05,
|
| 1168 |
+
"loss": 1.5011,
|
| 1169 |
+
"step": 3650
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 0.006126309393730352,
|
| 1173 |
+
"eval_loss": 1.5794349908828735,
|
| 1174 |
+
"eval_masked_accuracy": 0.6979591846466064,
|
| 1175 |
+
"eval_runtime": 1.7721,
|
| 1176 |
+
"eval_samples_per_second": 5.643,
|
| 1177 |
+
"eval_steps_per_second": 2.257,
|
| 1178 |
+
"step": 3650
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.006210231440219809,
|
| 1182 |
+
"grad_norm": 7.159238815307617,
|
| 1183 |
+
"learning_rate": 4.968965627208199e-05,
|
| 1184 |
+
"loss": 1.6134,
|
| 1185 |
+
"step": 3700
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.006210231440219809,
|
| 1189 |
+
"eval_loss": 1.4294860363006592,
|
| 1190 |
+
"eval_masked_accuracy": 0.7136752009391785,
|
| 1191 |
+
"eval_runtime": 1.752,
|
| 1192 |
+
"eval_samples_per_second": 5.708,
|
| 1193 |
+
"eval_steps_per_second": 2.283,
|
| 1194 |
+
"step": 3700
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 0.006294153486709265,
|
| 1198 |
+
"grad_norm": 5.560455799102783,
|
| 1199 |
+
"learning_rate": 4.968546016975752e-05,
|
| 1200 |
+
"loss": 1.5097,
|
| 1201 |
+
"step": 3750
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 0.006294153486709265,
|
| 1205 |
+
"eval_loss": 1.9169464111328125,
|
| 1206 |
+
"eval_masked_accuracy": 0.6929824352264404,
|
| 1207 |
+
"eval_runtime": 1.7457,
|
| 1208 |
+
"eval_samples_per_second": 5.728,
|
| 1209 |
+
"eval_steps_per_second": 2.291,
|
| 1210 |
+
"step": 3750
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.0063780755331987226,
|
| 1214 |
+
"grad_norm": 5.439815998077393,
|
| 1215 |
+
"learning_rate": 4.968126406743305e-05,
|
| 1216 |
+
"loss": 1.6706,
|
| 1217 |
+
"step": 3800
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 0.0063780755331987226,
|
| 1221 |
+
"eval_loss": 1.622685194015503,
|
| 1222 |
+
"eval_masked_accuracy": 0.6913580298423767,
|
| 1223 |
+
"eval_runtime": 1.7518,
|
| 1224 |
+
"eval_samples_per_second": 5.709,
|
| 1225 |
+
"eval_steps_per_second": 2.283,
|
| 1226 |
+
"step": 3800
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 0.006461997579688179,
|
| 1230 |
+
"grad_norm": 4.242193698883057,
|
| 1231 |
+
"learning_rate": 4.967706796510857e-05,
|
| 1232 |
+
"loss": 1.5511,
|
| 1233 |
+
"step": 3850
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 0.006461997579688179,
|
| 1237 |
+
"eval_loss": 1.3621394634246826,
|
| 1238 |
+
"eval_masked_accuracy": 0.7379912734031677,
|
| 1239 |
+
"eval_runtime": 1.7356,
|
| 1240 |
+
"eval_samples_per_second": 5.762,
|
| 1241 |
+
"eval_steps_per_second": 2.305,
|
| 1242 |
+
"step": 3850
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.0065459196261776365,
|
| 1246 |
+
"grad_norm": 5.056567668914795,
|
| 1247 |
+
"learning_rate": 4.96728718627841e-05,
|
| 1248 |
+
"loss": 1.6108,
|
| 1249 |
+
"step": 3900
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.0065459196261776365,
|
| 1253 |
+
"eval_loss": 1.5381476879119873,
|
| 1254 |
+
"eval_masked_accuracy": 0.7165354490280151,
|
| 1255 |
+
"eval_runtime": 1.7342,
|
| 1256 |
+
"eval_samples_per_second": 5.767,
|
| 1257 |
+
"eval_steps_per_second": 2.307,
|
| 1258 |
+
"step": 3900
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 0.006629841672667093,
|
| 1262 |
+
"grad_norm": 5.566115856170654,
|
| 1263 |
+
"learning_rate": 4.966867576045963e-05,
|
| 1264 |
+
"loss": 1.5858,
|
| 1265 |
+
"step": 3950
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 0.006629841672667093,
|
| 1269 |
+
"eval_loss": 1.9895532131195068,
|
| 1270 |
+
"eval_masked_accuracy": 0.6399999856948853,
|
| 1271 |
+
"eval_runtime": 1.7417,
|
| 1272 |
+
"eval_samples_per_second": 5.742,
|
| 1273 |
+
"eval_steps_per_second": 2.297,
|
| 1274 |
+
"step": 3950
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 0.00671376371915655,
|
| 1278 |
+
"grad_norm": 7.843978404998779,
|
| 1279 |
+
"learning_rate": 4.9664479658135155e-05,
|
| 1280 |
+
"loss": 1.5999,
|
| 1281 |
+
"step": 4000
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 0.00671376371915655,
|
| 1285 |
+
"eval_loss": 1.589036464691162,
|
| 1286 |
+
"eval_masked_accuracy": 0.6991525292396545,
|
| 1287 |
+
"eval_runtime": 1.7452,
|
| 1288 |
+
"eval_samples_per_second": 5.73,
|
| 1289 |
+
"eval_steps_per_second": 2.292,
|
| 1290 |
+
"step": 4000
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.006797685765646007,
|
| 1294 |
+
"grad_norm": 6.600104331970215,
|
| 1295 |
+
"learning_rate": 4.9660283555810683e-05,
|
| 1296 |
+
"loss": 1.6444,
|
| 1297 |
+
"step": 4050
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.006797685765646007,
|
| 1301 |
+
"eval_loss": 1.590256929397583,
|
| 1302 |
+
"eval_masked_accuracy": 0.6895161271095276,
|
| 1303 |
+
"eval_runtime": 1.7389,
|
| 1304 |
+
"eval_samples_per_second": 5.751,
|
| 1305 |
+
"eval_steps_per_second": 2.3,
|
| 1306 |
+
"step": 4050
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.006881607812135464,
|
| 1310 |
+
"grad_norm": 6.0659589767456055,
|
| 1311 |
+
"learning_rate": 4.965608745348621e-05,
|
| 1312 |
+
"loss": 1.5554,
|
| 1313 |
+
"step": 4100
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 0.006881607812135464,
|
| 1317 |
+
"eval_loss": 1.8275972604751587,
|
| 1318 |
+
"eval_masked_accuracy": 0.6558139324188232,
|
| 1319 |
+
"eval_runtime": 1.7513,
|
| 1320 |
+
"eval_samples_per_second": 5.71,
|
| 1321 |
+
"eval_steps_per_second": 2.284,
|
| 1322 |
+
"step": 4100
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 0.00696552985862492,
|
| 1326 |
+
"grad_norm": 6.09676456451416,
|
| 1327 |
+
"learning_rate": 4.9651891351161734e-05,
|
| 1328 |
+
"loss": 1.7191,
|
| 1329 |
+
"step": 4150
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 0.00696552985862492,
|
| 1333 |
+
"eval_loss": 1.8767850399017334,
|
| 1334 |
+
"eval_masked_accuracy": 0.6508620977401733,
|
| 1335 |
+
"eval_runtime": 1.7474,
|
| 1336 |
+
"eval_samples_per_second": 5.723,
|
| 1337 |
+
"eval_steps_per_second": 2.289,
|
| 1338 |
+
"step": 4150
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 0.0070494519051143775,
|
| 1342 |
+
"grad_norm": 5.208311080932617,
|
| 1343 |
+
"learning_rate": 4.964769524883726e-05,
|
| 1344 |
+
"loss": 1.585,
|
| 1345 |
+
"step": 4200
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 0.0070494519051143775,
|
| 1349 |
+
"eval_loss": 1.3652145862579346,
|
| 1350 |
+
"eval_masked_accuracy": 0.7037037014961243,
|
| 1351 |
+
"eval_runtime": 1.7463,
|
| 1352 |
+
"eval_samples_per_second": 5.726,
|
| 1353 |
+
"eval_steps_per_second": 2.291,
|
| 1354 |
+
"step": 4200
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.007133373951603834,
|
| 1358 |
+
"grad_norm": 8.517348289489746,
|
| 1359 |
+
"learning_rate": 4.964349914651279e-05,
|
| 1360 |
+
"loss": 1.6888,
|
| 1361 |
+
"step": 4250
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.007133373951603834,
|
| 1365 |
+
"eval_loss": 1.347320318222046,
|
| 1366 |
+
"eval_masked_accuracy": 0.7190082669258118,
|
| 1367 |
+
"eval_runtime": 1.7446,
|
| 1368 |
+
"eval_samples_per_second": 5.732,
|
| 1369 |
+
"eval_steps_per_second": 2.293,
|
| 1370 |
+
"step": 4250
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 0.0072172959980932915,
|
| 1374 |
+
"grad_norm": 5.57391357421875,
|
| 1375 |
+
"learning_rate": 4.963930304418832e-05,
|
| 1376 |
+
"loss": 1.6351,
|
| 1377 |
+
"step": 4300
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 0.0072172959980932915,
|
| 1381 |
+
"eval_loss": 1.563398003578186,
|
| 1382 |
+
"eval_masked_accuracy": 0.6952789425849915,
|
| 1383 |
+
"eval_runtime": 1.7535,
|
| 1384 |
+
"eval_samples_per_second": 5.703,
|
| 1385 |
+
"eval_steps_per_second": 2.281,
|
| 1386 |
+
"step": 4300
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 0.007301218044582748,
|
| 1390 |
+
"grad_norm": 4.073302745819092,
|
| 1391 |
+
"learning_rate": 4.963510694186385e-05,
|
| 1392 |
+
"loss": 1.7031,
|
| 1393 |
+
"step": 4350
|
| 1394 |
+
},
|
| 1395 |
+
{
|
| 1396 |
+
"epoch": 0.007301218044582748,
|
| 1397 |
+
"eval_loss": 1.7390921115875244,
|
| 1398 |
+
"eval_masked_accuracy": 0.6963562965393066,
|
| 1399 |
+
"eval_runtime": 1.8598,
|
| 1400 |
+
"eval_samples_per_second": 5.377,
|
| 1401 |
+
"eval_steps_per_second": 2.151,
|
| 1402 |
+
"step": 4350
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.007385140091072205,
|
| 1406 |
+
"grad_norm": 4.129016876220703,
|
| 1407 |
+
"learning_rate": 4.963091083953937e-05,
|
| 1408 |
+
"loss": 1.5611,
|
| 1409 |
+
"step": 4400
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.007385140091072205,
|
| 1413 |
+
"eval_loss": 1.7892725467681885,
|
| 1414 |
+
"eval_masked_accuracy": 0.7015503644943237,
|
| 1415 |
+
"eval_runtime": 1.7481,
|
| 1416 |
+
"eval_samples_per_second": 5.72,
|
| 1417 |
+
"eval_steps_per_second": 2.288,
|
| 1418 |
+
"step": 4400
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 0.007469062137561662,
|
| 1422 |
+
"grad_norm": 8.45355224609375,
|
| 1423 |
+
"learning_rate": 4.96267147372149e-05,
|
| 1424 |
+
"loss": 1.679,
|
| 1425 |
+
"step": 4450
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 0.007469062137561662,
|
| 1429 |
+
"eval_loss": 1.8994945287704468,
|
| 1430 |
+
"eval_masked_accuracy": 0.6711111068725586,
|
| 1431 |
+
"eval_runtime": 1.748,
|
| 1432 |
+
"eval_samples_per_second": 5.721,
|
| 1433 |
+
"eval_steps_per_second": 2.288,
|
| 1434 |
+
"step": 4450
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 0.0075529841840511185,
|
| 1438 |
+
"grad_norm": 7.353001594543457,
|
| 1439 |
+
"learning_rate": 4.9622602556936916e-05,
|
| 1440 |
+
"loss": 1.5084,
|
| 1441 |
+
"step": 4500
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 0.0075529841840511185,
|
| 1445 |
+
"eval_loss": 1.6633514165878296,
|
| 1446 |
+
"eval_masked_accuracy": 0.6792452931404114,
|
| 1447 |
+
"eval_runtime": 1.7365,
|
| 1448 |
+
"eval_samples_per_second": 5.759,
|
| 1449 |
+
"eval_steps_per_second": 2.303,
|
| 1450 |
+
"step": 4500
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 0.007636906230540575,
|
| 1454 |
+
"grad_norm": 5.420140266418457,
|
| 1455 |
+
"learning_rate": 4.9618406454612445e-05,
|
| 1456 |
+
"loss": 1.6768,
|
| 1457 |
+
"step": 4550
|
| 1458 |
+
},
|
| 1459 |
+
{
|
| 1460 |
+
"epoch": 0.007636906230540575,
|
| 1461 |
+
"eval_loss": 1.6823314428329468,
|
| 1462 |
+
"eval_masked_accuracy": 0.700421929359436,
|
| 1463 |
+
"eval_runtime": 1.7456,
|
| 1464 |
+
"eval_samples_per_second": 5.729,
|
| 1465 |
+
"eval_steps_per_second": 2.291,
|
| 1466 |
+
"step": 4550
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.0077208282770300325,
|
| 1470 |
+
"grad_norm": 5.6282572746276855,
|
| 1471 |
+
"learning_rate": 4.961421035228797e-05,
|
| 1472 |
+
"loss": 1.5346,
|
| 1473 |
+
"step": 4600
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.0077208282770300325,
|
| 1477 |
+
"eval_loss": 2.210347890853882,
|
| 1478 |
+
"eval_masked_accuracy": 0.6339285969734192,
|
| 1479 |
+
"eval_runtime": 1.7553,
|
| 1480 |
+
"eval_samples_per_second": 5.697,
|
| 1481 |
+
"eval_steps_per_second": 2.279,
|
| 1482 |
+
"step": 4600
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 0.007804750323519489,
|
| 1486 |
+
"grad_norm": 7.358382701873779,
|
| 1487 |
+
"learning_rate": 4.96100142499635e-05,
|
| 1488 |
+
"loss": 1.6792,
|
| 1489 |
+
"step": 4650
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 0.007804750323519489,
|
| 1493 |
+
"eval_loss": 1.742630958557129,
|
| 1494 |
+
"eval_masked_accuracy": 0.6728110313415527,
|
| 1495 |
+
"eval_runtime": 1.7331,
|
| 1496 |
+
"eval_samples_per_second": 5.77,
|
| 1497 |
+
"eval_steps_per_second": 2.308,
|
| 1498 |
+
"step": 4650
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 0.007888672370008946,
|
| 1502 |
+
"grad_norm": 5.980144500732422,
|
| 1503 |
+
"learning_rate": 4.960581814763902e-05,
|
| 1504 |
+
"loss": 1.4871,
|
| 1505 |
+
"step": 4700
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 0.007888672370008946,
|
| 1509 |
+
"eval_loss": 1.4571318626403809,
|
| 1510 |
+
"eval_masked_accuracy": 0.7166666388511658,
|
| 1511 |
+
"eval_runtime": 1.7531,
|
| 1512 |
+
"eval_samples_per_second": 5.704,
|
| 1513 |
+
"eval_steps_per_second": 2.282,
|
| 1514 |
+
"step": 4700
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.007972594416498403,
|
| 1518 |
+
"grad_norm": 8.18883228302002,
|
| 1519 |
+
"learning_rate": 4.960162204531455e-05,
|
| 1520 |
+
"loss": 1.527,
|
| 1521 |
+
"step": 4750
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.007972594416498403,
|
| 1525 |
+
"eval_loss": 2.062413454055786,
|
| 1526 |
+
"eval_masked_accuracy": 0.6695278882980347,
|
| 1527 |
+
"eval_runtime": 1.748,
|
| 1528 |
+
"eval_samples_per_second": 5.721,
|
| 1529 |
+
"eval_steps_per_second": 2.288,
|
| 1530 |
+
"step": 4750
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 0.00805651646298786,
|
| 1534 |
+
"grad_norm": 4.835183143615723,
|
| 1535 |
+
"learning_rate": 4.959742594299008e-05,
|
| 1536 |
+
"loss": 1.591,
|
| 1537 |
+
"step": 4800
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 0.00805651646298786,
|
| 1541 |
+
"eval_loss": 1.690118432044983,
|
| 1542 |
+
"eval_masked_accuracy": 0.7049180269241333,
|
| 1543 |
+
"eval_runtime": 1.7383,
|
| 1544 |
+
"eval_samples_per_second": 5.753,
|
| 1545 |
+
"eval_steps_per_second": 2.301,
|
| 1546 |
+
"step": 4800
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 0.008140438509477316,
|
| 1550 |
+
"grad_norm": 5.039312362670898,
|
| 1551 |
+
"learning_rate": 4.959322984066561e-05,
|
| 1552 |
+
"loss": 1.5386,
|
| 1553 |
+
"step": 4850
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 0.008140438509477316,
|
| 1557 |
+
"eval_loss": 1.9135382175445557,
|
| 1558 |
+
"eval_masked_accuracy": 0.6181818246841431,
|
| 1559 |
+
"eval_runtime": 1.7445,
|
| 1560 |
+
"eval_samples_per_second": 5.732,
|
| 1561 |
+
"eval_steps_per_second": 2.293,
|
| 1562 |
+
"step": 4850
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 0.008224360555966774,
|
| 1566 |
+
"grad_norm": 6.3293890953063965,
|
| 1567 |
+
"learning_rate": 4.958903373834114e-05,
|
| 1568 |
+
"loss": 1.4752,
|
| 1569 |
+
"step": 4900
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 0.008224360555966774,
|
| 1573 |
+
"eval_loss": 1.6353566646575928,
|
| 1574 |
+
"eval_masked_accuracy": 0.7319999933242798,
|
| 1575 |
+
"eval_runtime": 1.8458,
|
| 1576 |
+
"eval_samples_per_second": 5.418,
|
| 1577 |
+
"eval_steps_per_second": 2.167,
|
| 1578 |
+
"step": 4900
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.008308282602456231,
|
| 1582 |
+
"grad_norm": 7.455787658691406,
|
| 1583 |
+
"learning_rate": 4.958483763601666e-05,
|
| 1584 |
+
"loss": 1.5304,
|
| 1585 |
+
"step": 4950
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.008308282602456231,
|
| 1589 |
+
"eval_loss": 1.8691352605819702,
|
| 1590 |
+
"eval_masked_accuracy": 0.6653386354446411,
|
| 1591 |
+
"eval_runtime": 1.7533,
|
| 1592 |
+
"eval_samples_per_second": 5.704,
|
| 1593 |
+
"eval_steps_per_second": 2.281,
|
| 1594 |
+
"step": 4950
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 0.008392204648945687,
|
| 1598 |
+
"grad_norm": 5.682205677032471,
|
| 1599 |
+
"learning_rate": 4.958064153369219e-05,
|
| 1600 |
+
"loss": 1.5945,
|
| 1601 |
+
"step": 5000
|
| 1602 |
+
},
|
| 1603 |
+
{
|
| 1604 |
+
"epoch": 0.008392204648945687,
|
| 1605 |
+
"eval_loss": 1.6161428689956665,
|
| 1606 |
+
"eval_masked_accuracy": 0.6964285969734192,
|
| 1607 |
+
"eval_runtime": 1.7814,
|
| 1608 |
+
"eval_samples_per_second": 5.614,
|
| 1609 |
+
"eval_steps_per_second": 2.245,
|
| 1610 |
+
"step": 5000
|
| 1611 |
+
}
|
| 1612 |
+
],
|
| 1613 |
+
"logging_steps": 50,
|
| 1614 |
+
"max_steps": 595791,
|
| 1615 |
+
"num_input_tokens_seen": 0,
|
| 1616 |
+
"num_train_epochs": 1,
|
| 1617 |
+
"save_steps": 5000,
|
| 1618 |
+
"stateful_callbacks": {
|
| 1619 |
+
"TrainerControl": {
|
| 1620 |
+
"args": {
|
| 1621 |
+
"should_epoch_stop": false,
|
| 1622 |
+
"should_evaluate": false,
|
| 1623 |
+
"should_log": false,
|
| 1624 |
+
"should_save": true,
|
| 1625 |
+
"should_training_stop": false
|
| 1626 |
+
},
|
| 1627 |
+
"attributes": {}
|
| 1628 |
+
}
|
| 1629 |
+
},
|
| 1630 |
+
"total_flos": 8043289130117280.0,
|
| 1631 |
+
"train_batch_size": 2,
|
| 1632 |
+
"trial_name": null,
|
| 1633 |
+
"trial_params": null
|
| 1634 |
+
}
|
last-checkpoint/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6001e233381881441a26ca3b9298910952fc67e1181629bb9eb549c5280ee232
|
| 3 |
+
size 5240
|