Upload folder using huggingface_hub
Browse files- checkpoint-3500/config.json +16 -0
- checkpoint-3500/generation_config.json +6 -0
- checkpoint-3500/model.safetensors +3 -0
- checkpoint-3500/optimizer.pt +3 -0
- checkpoint-3500/rng_state.pth +3 -0
- checkpoint-3500/scaler.pt +3 -0
- checkpoint-3500/scheduler.pt +3 -0
- checkpoint-3500/trainer_state.json +2484 -0
- checkpoint-3500/training_args.bin +3 -0
checkpoint-3500/config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"FastHyperNet125mForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "float32",
|
| 6 |
+
"hidden_size": 768,
|
| 7 |
+
"hyperbolic_dim": 64,
|
| 8 |
+
"intermediate_size": 2048,
|
| 9 |
+
"max_position_embeddings": 512,
|
| 10 |
+
"model_type": "fasthypernet_125m_tagless",
|
| 11 |
+
"num_attention_heads": 12,
|
| 12 |
+
"num_hidden_layers": 16,
|
| 13 |
+
"transformers_version": "5.13.1",
|
| 14 |
+
"use_cache": false,
|
| 15 |
+
"vocab_size": 8192
|
| 16 |
+
}
|
checkpoint-3500/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"output_attentions": false,
|
| 4 |
+
"output_hidden_states": false,
|
| 5 |
+
"transformers_version": "5.13.1"
|
| 6 |
+
}
|
checkpoint-3500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a5e3b063b440e901eaee4eb59ce7038b3d2f19eb0d02bbc4ab9ab9eb13126c54
|
| 3 |
+
size 542849784
|
checkpoint-3500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:01caff722d90d1227e10a744c5386d0998553bd7e20dbcb913f4b8f366cf2ff8
|
| 3 |
+
size 273770959
|
checkpoint-3500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8217f5230fbec97d48feabba8b8e71534f9e3911332eacd817bf19ff6ffad57a
|
| 3 |
+
size 14645
|
checkpoint-3500/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e0214163a660267ec874ade1dadf3d522c85a3bab21646f9de4125be22fa135
|
| 3 |
+
size 1383
|
checkpoint-3500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d31120910c5d761ad50ccd3a75a4671399d42140a71899d2d942423c29e241c7
|
| 3 |
+
size 1465
|
checkpoint-3500/trainer_state.json
ADDED
|
@@ -0,0 +1,2484 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.1213823718115452,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 3500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00034680677660441485,
|
| 14 |
+
"grad_norm": 8.286304473876953,
|
| 15 |
+
"learning_rate": 2.6999999999999996e-06,
|
| 16 |
+
"loss": 36.023321533203124,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0006936135532088297,
|
| 21 |
+
"grad_norm": 5.303624153137207,
|
| 22 |
+
"learning_rate": 5.7e-06,
|
| 23 |
+
"loss": 35.81325988769531,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0010404203298132447,
|
| 28 |
+
"grad_norm": 14.0803861618042,
|
| 29 |
+
"learning_rate": 8.7e-06,
|
| 30 |
+
"loss": 35.61415100097656,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0013872271064176594,
|
| 35 |
+
"grad_norm": 42.10546875,
|
| 36 |
+
"learning_rate": 1.17e-05,
|
| 37 |
+
"loss": 35.36910400390625,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0017340338830220744,
|
| 42 |
+
"grad_norm": 43.37276840209961,
|
| 43 |
+
"learning_rate": 1.47e-05,
|
| 44 |
+
"loss": 34.86077880859375,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0020808406596264893,
|
| 49 |
+
"grad_norm": Infinity,
|
| 50 |
+
"learning_rate": 1.7699999999999997e-05,
|
| 51 |
+
"loss": 34.67452697753906,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.002427647436230904,
|
| 56 |
+
"grad_norm": 86.09446716308594,
|
| 57 |
+
"learning_rate": 2.04e-05,
|
| 58 |
+
"loss": 34.29078979492188,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.002774454212835319,
|
| 63 |
+
"grad_norm": 43.50123977661133,
|
| 64 |
+
"learning_rate": 2.34e-05,
|
| 65 |
+
"loss": 33.816766357421876,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0031212609894397335,
|
| 70 |
+
"grad_norm": 104.7258529663086,
|
| 71 |
+
"learning_rate": 2.6399999999999995e-05,
|
| 72 |
+
"loss": 33.36159057617188,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0034680677660441487,
|
| 77 |
+
"grad_norm": 66.46515655517578,
|
| 78 |
+
"learning_rate": 2.91e-05,
|
| 79 |
+
"loss": 33.181732177734375,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.0038148745426485635,
|
| 84 |
+
"grad_norm": 127.66901397705078,
|
| 85 |
+
"learning_rate": 3.2099999999999994e-05,
|
| 86 |
+
"loss": 32.83320007324219,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.004161681319252979,
|
| 91 |
+
"grad_norm": 71.67156982421875,
|
| 92 |
+
"learning_rate": 3.51e-05,
|
| 93 |
+
"loss": 32.48929138183594,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.004508488095857393,
|
| 98 |
+
"grad_norm": 76.15655517578125,
|
| 99 |
+
"learning_rate": 3.81e-05,
|
| 100 |
+
"loss": 32.10602722167969,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.004855294872461808,
|
| 105 |
+
"grad_norm": 38.65349578857422,
|
| 106 |
+
"learning_rate": 4.11e-05,
|
| 107 |
+
"loss": 31.449526977539062,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.005202101649066222,
|
| 112 |
+
"grad_norm": 59.55622863769531,
|
| 113 |
+
"learning_rate": 4.4099999999999995e-05,
|
| 114 |
+
"loss": 31.01339111328125,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.005548908425670638,
|
| 119 |
+
"grad_norm": 130.26136779785156,
|
| 120 |
+
"learning_rate": 4.709999999999999e-05,
|
| 121 |
+
"loss": 30.92586669921875,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.005895715202275053,
|
| 126 |
+
"grad_norm": 77.2778549194336,
|
| 127 |
+
"learning_rate": 5.01e-05,
|
| 128 |
+
"loss": 30.492510986328124,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.006242521978879467,
|
| 133 |
+
"grad_norm": 29.711563110351562,
|
| 134 |
+
"learning_rate": 5.309999999999999e-05,
|
| 135 |
+
"loss": 29.836746215820312,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.006589328755483882,
|
| 140 |
+
"grad_norm": 29.891956329345703,
|
| 141 |
+
"learning_rate": 5.6099999999999995e-05,
|
| 142 |
+
"loss": 29.291275024414062,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.006936135532088297,
|
| 147 |
+
"grad_norm": 30.56305503845215,
|
| 148 |
+
"learning_rate": 5.91e-05,
|
| 149 |
+
"loss": 29.14805908203125,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.007282942308692712,
|
| 154 |
+
"grad_norm": 29.489208221435547,
|
| 155 |
+
"learning_rate": 6.209999999999999e-05,
|
| 156 |
+
"loss": 28.80553283691406,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.007629749085297127,
|
| 161 |
+
"grad_norm": 37.147377014160156,
|
| 162 |
+
"learning_rate": 6.51e-05,
|
| 163 |
+
"loss": 28.694085693359376,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.007976555861901541,
|
| 168 |
+
"grad_norm": 28.496992111206055,
|
| 169 |
+
"learning_rate": 6.81e-05,
|
| 170 |
+
"loss": 28.166943359375,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.008323362638505957,
|
| 175 |
+
"grad_norm": 29.42328453063965,
|
| 176 |
+
"learning_rate": 7.11e-05,
|
| 177 |
+
"loss": 27.878482055664062,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.008670169415110372,
|
| 182 |
+
"grad_norm": 27.867637634277344,
|
| 183 |
+
"learning_rate": 7.41e-05,
|
| 184 |
+
"loss": 26.99749755859375,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.009016976191714786,
|
| 189 |
+
"grad_norm": 29.695758819580078,
|
| 190 |
+
"learning_rate": 7.709999999999999e-05,
|
| 191 |
+
"loss": 27.208102416992187,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.0093637829683192,
|
| 196 |
+
"grad_norm": 34.605934143066406,
|
| 197 |
+
"learning_rate": 8.01e-05,
|
| 198 |
+
"loss": 26.798681640625,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.009710589744923616,
|
| 203 |
+
"grad_norm": 35.05394744873047,
|
| 204 |
+
"learning_rate": 8.31e-05,
|
| 205 |
+
"loss": 27.192486572265626,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.01005739652152803,
|
| 210 |
+
"grad_norm": 19.137067794799805,
|
| 211 |
+
"learning_rate": 8.609999999999999e-05,
|
| 212 |
+
"loss": 26.89140930175781,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.010404203298132445,
|
| 217 |
+
"grad_norm": 24.543006896972656,
|
| 218 |
+
"learning_rate": 8.879999999999999e-05,
|
| 219 |
+
"loss": 26.167477416992188,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.010751010074736861,
|
| 224 |
+
"grad_norm": 18.849651336669922,
|
| 225 |
+
"learning_rate": 9.18e-05,
|
| 226 |
+
"loss": 26.41388854980469,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.011097816851341275,
|
| 231 |
+
"grad_norm": 21.179908752441406,
|
| 232 |
+
"learning_rate": 9.479999999999999e-05,
|
| 233 |
+
"loss": 26.003546142578124,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.01144462362794569,
|
| 238 |
+
"grad_norm": 19.344528198242188,
|
| 239 |
+
"learning_rate": 9.779999999999999e-05,
|
| 240 |
+
"loss": 26.271051025390626,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.011791430404550106,
|
| 245 |
+
"grad_norm": 19.74913215637207,
|
| 246 |
+
"learning_rate": 0.0001008,
|
| 247 |
+
"loss": 26.064590454101562,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.01213823718115452,
|
| 252 |
+
"grad_norm": 17.638103485107422,
|
| 253 |
+
"learning_rate": 0.00010379999999999999,
|
| 254 |
+
"loss": 25.863119506835936,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.012485043957758934,
|
| 259 |
+
"grad_norm": 24.873981475830078,
|
| 260 |
+
"learning_rate": 0.00010679999999999998,
|
| 261 |
+
"loss": 26.28948059082031,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.01283185073436335,
|
| 266 |
+
"grad_norm": 17.048828125,
|
| 267 |
+
"learning_rate": 0.00010979999999999999,
|
| 268 |
+
"loss": 26.215179443359375,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.013178657510967764,
|
| 273 |
+
"grad_norm": 21.703975677490234,
|
| 274 |
+
"learning_rate": 0.00011279999999999999,
|
| 275 |
+
"loss": 26.152392578125,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.013525464287572179,
|
| 280 |
+
"grad_norm": 22.66498565673828,
|
| 281 |
+
"learning_rate": 0.0001158,
|
| 282 |
+
"loss": 26.0640625,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.013872271064176595,
|
| 287 |
+
"grad_norm": 16.945016860961914,
|
| 288 |
+
"learning_rate": 0.0001188,
|
| 289 |
+
"loss": 26.023715209960937,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.01421907784078101,
|
| 294 |
+
"grad_norm": 27.866268157958984,
|
| 295 |
+
"learning_rate": 0.00012179999999999999,
|
| 296 |
+
"loss": 25.43433380126953,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.014565884617385423,
|
| 301 |
+
"grad_norm": 22.481279373168945,
|
| 302 |
+
"learning_rate": 0.00012479999999999997,
|
| 303 |
+
"loss": 25.464639282226564,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.014912691393989838,
|
| 308 |
+
"grad_norm": 14.993969917297363,
|
| 309 |
+
"learning_rate": 0.0001278,
|
| 310 |
+
"loss": 25.218385314941408,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.015259498170594254,
|
| 315 |
+
"grad_norm": 14.426427841186523,
|
| 316 |
+
"learning_rate": 0.00013079999999999998,
|
| 317 |
+
"loss": 24.94810333251953,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.015606304947198668,
|
| 322 |
+
"grad_norm": 15.592818260192871,
|
| 323 |
+
"learning_rate": 0.0001338,
|
| 324 |
+
"loss": 25.38585662841797,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.015953111723803082,
|
| 329 |
+
"grad_norm": 14.709174156188965,
|
| 330 |
+
"learning_rate": 0.0001368,
|
| 331 |
+
"loss": 24.31098175048828,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.016299918500407497,
|
| 336 |
+
"grad_norm": 13.651994705200195,
|
| 337 |
+
"learning_rate": 0.00013979999999999998,
|
| 338 |
+
"loss": 24.82652893066406,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.016646725277011915,
|
| 343 |
+
"grad_norm": 17.521072387695312,
|
| 344 |
+
"learning_rate": 0.00014279999999999997,
|
| 345 |
+
"loss": 25.121923828125,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.01699353205361633,
|
| 350 |
+
"grad_norm": 13.458234786987305,
|
| 351 |
+
"learning_rate": 0.0001458,
|
| 352 |
+
"loss": 24.447775268554686,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.017340338830220743,
|
| 357 |
+
"grad_norm": 13.57589340209961,
|
| 358 |
+
"learning_rate": 0.00014879999999999998,
|
| 359 |
+
"loss": 24.76665802001953,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.017687145606825157,
|
| 364 |
+
"grad_norm": 12.934654235839844,
|
| 365 |
+
"learning_rate": 0.00015179999999999998,
|
| 366 |
+
"loss": 24.170372009277344,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.018033952383429572,
|
| 371 |
+
"grad_norm": 13.884750366210938,
|
| 372 |
+
"learning_rate": 0.0001548,
|
| 373 |
+
"loss": 24.240016174316406,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.018380759160033986,
|
| 378 |
+
"grad_norm": 13.660422325134277,
|
| 379 |
+
"learning_rate": 0.0001578,
|
| 380 |
+
"loss": 23.897187805175783,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.0187275659366384,
|
| 385 |
+
"grad_norm": 11.849883079528809,
|
| 386 |
+
"learning_rate": 0.0001608,
|
| 387 |
+
"loss": 23.574449157714845,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.019074372713242818,
|
| 392 |
+
"grad_norm": 10.595759391784668,
|
| 393 |
+
"learning_rate": 0.0001638,
|
| 394 |
+
"loss": 23.569731140136717,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.019421179489847232,
|
| 399 |
+
"grad_norm": 12.363926887512207,
|
| 400 |
+
"learning_rate": 0.0001668,
|
| 401 |
+
"loss": 23.617283630371094,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.019767986266451647,
|
| 406 |
+
"grad_norm": 10.968243598937988,
|
| 407 |
+
"learning_rate": 0.00016979999999999998,
|
| 408 |
+
"loss": 23.990586853027345,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.02011479304305606,
|
| 413 |
+
"grad_norm": 12.141168594360352,
|
| 414 |
+
"learning_rate": 0.00017279999999999997,
|
| 415 |
+
"loss": 23.807048034667968,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.020461599819660475,
|
| 420 |
+
"grad_norm": 14.928491592407227,
|
| 421 |
+
"learning_rate": 0.00017579999999999996,
|
| 422 |
+
"loss": 22.980938720703126,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.02080840659626489,
|
| 427 |
+
"grad_norm": 11.39072036743164,
|
| 428 |
+
"learning_rate": 0.00017879999999999998,
|
| 429 |
+
"loss": 23.521900939941407,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.021155213372869307,
|
| 434 |
+
"grad_norm": 10.157258033752441,
|
| 435 |
+
"learning_rate": 0.00018179999999999997,
|
| 436 |
+
"loss": 23.075851440429688,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.021502020149473722,
|
| 441 |
+
"grad_norm": 12.673665046691895,
|
| 442 |
+
"learning_rate": 0.0001848,
|
| 443 |
+
"loss": 23.70465850830078,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.021848826926078136,
|
| 448 |
+
"grad_norm": 11.242607116699219,
|
| 449 |
+
"learning_rate": 0.00018779999999999998,
|
| 450 |
+
"loss": 22.982296752929688,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.02219563370268255,
|
| 455 |
+
"grad_norm": 25.496898651123047,
|
| 456 |
+
"learning_rate": 0.00019079999999999998,
|
| 457 |
+
"loss": 22.516278076171876,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.022542440479286965,
|
| 462 |
+
"grad_norm": 11.215885162353516,
|
| 463 |
+
"learning_rate": 0.0001938,
|
| 464 |
+
"loss": 22.14655456542969,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.02288924725589138,
|
| 469 |
+
"grad_norm": 9.149953842163086,
|
| 470 |
+
"learning_rate": 0.00019679999999999999,
|
| 471 |
+
"loss": 22.114730834960938,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.023236054032495793,
|
| 476 |
+
"grad_norm": 14.604310035705566,
|
| 477 |
+
"learning_rate": 0.0001995,
|
| 478 |
+
"loss": 22.670735168457032,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.02358286080910021,
|
| 483 |
+
"grad_norm": 11.181646347045898,
|
| 484 |
+
"learning_rate": 0.0002025,
|
| 485 |
+
"loss": 22.34198455810547,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.023929667585704625,
|
| 490 |
+
"grad_norm": 10.59487533569336,
|
| 491 |
+
"learning_rate": 0.0002055,
|
| 492 |
+
"loss": 22.268771362304687,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.02427647436230904,
|
| 497 |
+
"grad_norm": 11.758952140808105,
|
| 498 |
+
"learning_rate": 0.00020849999999999997,
|
| 499 |
+
"loss": 21.350621032714844,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.024623281138913454,
|
| 504 |
+
"grad_norm": 18.59368896484375,
|
| 505 |
+
"learning_rate": 0.00021149999999999996,
|
| 506 |
+
"loss": 21.507183837890626,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.02497008791551787,
|
| 511 |
+
"grad_norm": 9.710185050964355,
|
| 512 |
+
"learning_rate": 0.00021449999999999998,
|
| 513 |
+
"loss": 21.502236938476564,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.025316894692122283,
|
| 518 |
+
"grad_norm": 20.68129539489746,
|
| 519 |
+
"learning_rate": 0.00021749999999999997,
|
| 520 |
+
"loss": 21.232615661621093,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.0256637014687267,
|
| 525 |
+
"grad_norm": 9.399124145507812,
|
| 526 |
+
"learning_rate": 0.00022049999999999997,
|
| 527 |
+
"loss": 21.146585083007814,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.026010508245331115,
|
| 532 |
+
"grad_norm": 9.145174026489258,
|
| 533 |
+
"learning_rate": 0.00022349999999999998,
|
| 534 |
+
"loss": 21.06390838623047,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.02635731502193553,
|
| 539 |
+
"grad_norm": 11.161269187927246,
|
| 540 |
+
"learning_rate": 0.00022649999999999998,
|
| 541 |
+
"loss": 20.896176147460938,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.026704121798539943,
|
| 546 |
+
"grad_norm": 8.362421035766602,
|
| 547 |
+
"learning_rate": 0.0002295,
|
| 548 |
+
"loss": 20.884358215332032,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.027050928575144358,
|
| 553 |
+
"grad_norm": 9.504199981689453,
|
| 554 |
+
"learning_rate": 0.00023249999999999999,
|
| 555 |
+
"loss": 20.086947631835937,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.027397735351748772,
|
| 560 |
+
"grad_norm": 8.848526954650879,
|
| 561 |
+
"learning_rate": 0.00023549999999999998,
|
| 562 |
+
"loss": 20.411599731445314,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.02774454212835319,
|
| 567 |
+
"grad_norm": 9.612874984741211,
|
| 568 |
+
"learning_rate": 0.0002385,
|
| 569 |
+
"loss": 20.284182739257812,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.028091348904957604,
|
| 574 |
+
"grad_norm": 8.688453674316406,
|
| 575 |
+
"learning_rate": 0.0002415,
|
| 576 |
+
"loss": 20.024198913574217,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.02843815568156202,
|
| 581 |
+
"grad_norm": 9.779630661010742,
|
| 582 |
+
"learning_rate": 0.0002445,
|
| 583 |
+
"loss": 19.91345977783203,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.028784962458166433,
|
| 588 |
+
"grad_norm": 12.37082576751709,
|
| 589 |
+
"learning_rate": 0.0002472,
|
| 590 |
+
"loss": 19.54875030517578,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.029131769234770847,
|
| 595 |
+
"grad_norm": 8.736369132995605,
|
| 596 |
+
"learning_rate": 0.00025019999999999996,
|
| 597 |
+
"loss": 19.499098205566405,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.02947857601137526,
|
| 602 |
+
"grad_norm": 10.752023696899414,
|
| 603 |
+
"learning_rate": 0.0002532,
|
| 604 |
+
"loss": 19.327198791503907,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.029825382787979676,
|
| 609 |
+
"grad_norm": 9.755398750305176,
|
| 610 |
+
"learning_rate": 0.0002562,
|
| 611 |
+
"loss": 19.291722106933594,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.030172189564584093,
|
| 616 |
+
"grad_norm": 8.066308975219727,
|
| 617 |
+
"learning_rate": 0.00025919999999999996,
|
| 618 |
+
"loss": 18.516694641113283,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.030518996341188508,
|
| 623 |
+
"grad_norm": 8.660676002502441,
|
| 624 |
+
"learning_rate": 0.0002622,
|
| 625 |
+
"loss": 18.70814666748047,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.030865803117792922,
|
| 630 |
+
"grad_norm": 8.439098358154297,
|
| 631 |
+
"learning_rate": 0.0002652,
|
| 632 |
+
"loss": 18.787985229492186,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.031212609894397336,
|
| 637 |
+
"grad_norm": 8.525120735168457,
|
| 638 |
+
"learning_rate": 0.00026819999999999996,
|
| 639 |
+
"loss": 18.356887817382812,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.031559416671001754,
|
| 644 |
+
"grad_norm": 9.237743377685547,
|
| 645 |
+
"learning_rate": 0.0002712,
|
| 646 |
+
"loss": 18.07917022705078,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.031906223447606165,
|
| 651 |
+
"grad_norm": 8.137825965881348,
|
| 652 |
+
"learning_rate": 0.0002742,
|
| 653 |
+
"loss": 18.066172790527343,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.03225303022421058,
|
| 658 |
+
"grad_norm": 20.67136001586914,
|
| 659 |
+
"learning_rate": 0.0002772,
|
| 660 |
+
"loss": 18.19664306640625,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.032599837000814993,
|
| 665 |
+
"grad_norm": 8.68689250946045,
|
| 666 |
+
"learning_rate": 0.0002802,
|
| 667 |
+
"loss": 17.834249877929686,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.03294664377741941,
|
| 672 |
+
"grad_norm": 8.076814651489258,
|
| 673 |
+
"learning_rate": 0.00028319999999999994,
|
| 674 |
+
"loss": 17.645645141601562,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.03329345055402383,
|
| 679 |
+
"grad_norm": 8.278287887573242,
|
| 680 |
+
"learning_rate": 0.00028619999999999996,
|
| 681 |
+
"loss": 17.20880889892578,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.03364025733062824,
|
| 686 |
+
"grad_norm": 8.195657730102539,
|
| 687 |
+
"learning_rate": 0.0002892,
|
| 688 |
+
"loss": 17.02202453613281,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.03398706410723266,
|
| 693 |
+
"grad_norm": 35.29956817626953,
|
| 694 |
+
"learning_rate": 0.00029219999999999995,
|
| 695 |
+
"loss": 17.46351013183594,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.03433387088383707,
|
| 700 |
+
"grad_norm": 14.36285400390625,
|
| 701 |
+
"learning_rate": 0.00029519999999999997,
|
| 702 |
+
"loss": 17.128692626953125,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.034680677660441486,
|
| 707 |
+
"grad_norm": 117.96296691894531,
|
| 708 |
+
"learning_rate": 0.0002979,
|
| 709 |
+
"loss": 16.577227783203124,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.0350274844370459,
|
| 714 |
+
"grad_norm": 33.18951416015625,
|
| 715 |
+
"learning_rate": 0.00029999999932027514,
|
| 716 |
+
"loss": 16.93028106689453,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.035374291213650315,
|
| 721 |
+
"grad_norm": 16.420063018798828,
|
| 722 |
+
"learning_rate": 0.00029999998723627847,
|
| 723 |
+
"loss": 17.68489990234375,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.03572109799025473,
|
| 728 |
+
"grad_norm": 37.7625732421875,
|
| 729 |
+
"learning_rate": 0.0002999999600472873,
|
| 730 |
+
"loss": 17.295022583007814,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.036067904766859143,
|
| 735 |
+
"grad_norm": 32.17815399169922,
|
| 736 |
+
"learning_rate": 0.00029999992266242694,
|
| 737 |
+
"loss": 17.40197296142578,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.03641471154346356,
|
| 742 |
+
"grad_norm": 10.255291938781738,
|
| 743 |
+
"learning_rate": 0.00029999986677395476,
|
| 744 |
+
"loss": 17.136293029785158,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.03676151832006797,
|
| 749 |
+
"grad_norm": 7.522293567657471,
|
| 750 |
+
"learning_rate": 0.0002999997957805001,
|
| 751 |
+
"loss": 16.333123779296876,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.03710832509667239,
|
| 756 |
+
"grad_norm": 7.725010395050049,
|
| 757 |
+
"learning_rate": 0.00029999970968207007,
|
| 758 |
+
"loss": 15.974671936035156,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.0374551318732768,
|
| 763 |
+
"grad_norm": 7.349689483642578,
|
| 764 |
+
"learning_rate": 0.00029999960847867344,
|
| 765 |
+
"loss": 15.520753479003906,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.03780193864988122,
|
| 770 |
+
"grad_norm": 6.910863399505615,
|
| 771 |
+
"learning_rate": 0.00029999949217032024,
|
| 772 |
+
"loss": 15.53778533935547,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.038148745426485636,
|
| 777 |
+
"grad_norm": 7.105229377746582,
|
| 778 |
+
"learning_rate": 0.00029999936075702236,
|
| 779 |
+
"loss": 15.192759704589843,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.03849555220309005,
|
| 784 |
+
"grad_norm": 6.765523910522461,
|
| 785 |
+
"learning_rate": 0.0002999992142387929,
|
| 786 |
+
"loss": 15.362620544433593,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.038842358979694465,
|
| 791 |
+
"grad_norm": 7.596284866333008,
|
| 792 |
+
"learning_rate": 0.0002999990526156467,
|
| 793 |
+
"loss": 15.27081298828125,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.039189165756298876,
|
| 798 |
+
"grad_norm": 7.164301872253418,
|
| 799 |
+
"learning_rate": 0.0002999988758876,
|
| 800 |
+
"loss": 14.56300048828125,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.039535972532903293,
|
| 805 |
+
"grad_norm": 6.729367256164551,
|
| 806 |
+
"learning_rate": 0.0002999986840546706,
|
| 807 |
+
"loss": 14.51199951171875,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.039882779309507704,
|
| 812 |
+
"grad_norm": 6.131908893585205,
|
| 813 |
+
"learning_rate": 0.00029999847711687785,
|
| 814 |
+
"loss": 14.647183227539063,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.04022958608611212,
|
| 819 |
+
"grad_norm": 7.322798728942871,
|
| 820 |
+
"learning_rate": 0.00029999825507424254,
|
| 821 |
+
"loss": 14.303775024414062,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.04057639286271654,
|
| 826 |
+
"grad_norm": 7.279247760772705,
|
| 827 |
+
"learning_rate": 0.00029999801792678706,
|
| 828 |
+
"loss": 14.378622436523438,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.04092319963932095,
|
| 833 |
+
"grad_norm": 6.741374492645264,
|
| 834 |
+
"learning_rate": 0.00029999776567453526,
|
| 835 |
+
"loss": 13.964057922363281,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.04127000641592537,
|
| 840 |
+
"grad_norm": 6.461069583892822,
|
| 841 |
+
"learning_rate": 0.00029999749831751256,
|
| 842 |
+
"loss": 14.041261291503906,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.04161681319252978,
|
| 847 |
+
"grad_norm": 6.411841869354248,
|
| 848 |
+
"learning_rate": 0.0002999972158557459,
|
| 849 |
+
"loss": 13.568597412109375,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.0419636199691342,
|
| 854 |
+
"grad_norm": 6.214684963226318,
|
| 855 |
+
"learning_rate": 0.00029999691828926366,
|
| 856 |
+
"loss": 13.064254760742188,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.042310426745738615,
|
| 861 |
+
"grad_norm": 6.115879058837891,
|
| 862 |
+
"learning_rate": 0.0002999966056180959,
|
| 863 |
+
"loss": 14.200587463378906,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.042657233522343026,
|
| 868 |
+
"grad_norm": 6.505321979522705,
|
| 869 |
+
"learning_rate": 0.00029999627784227406,
|
| 870 |
+
"loss": 13.873959350585938,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.043004040298947444,
|
| 875 |
+
"grad_norm": 6.690847873687744,
|
| 876 |
+
"learning_rate": 0.0002999959349618311,
|
| 877 |
+
"loss": 13.29089813232422,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.043350847075551854,
|
| 882 |
+
"grad_norm": 6.375129699707031,
|
| 883 |
+
"learning_rate": 0.0002999955769768017,
|
| 884 |
+
"loss": 13.220611572265625,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.04369765385215627,
|
| 889 |
+
"grad_norm": 5.927398204803467,
|
| 890 |
+
"learning_rate": 0.00029999520388722167,
|
| 891 |
+
"loss": 13.184442138671875,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.04404446062876068,
|
| 896 |
+
"grad_norm": 6.170562267303467,
|
| 897 |
+
"learning_rate": 0.00029999481569312875,
|
| 898 |
+
"loss": 12.594881439208985,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.0443912674053651,
|
| 903 |
+
"grad_norm": 7.415892601013184,
|
| 904 |
+
"learning_rate": 0.000299994412394562,
|
| 905 |
+
"loss": 12.665945434570313,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.04473807418196952,
|
| 910 |
+
"grad_norm": 6.139370441436768,
|
| 911 |
+
"learning_rate": 0.0002999939939915621,
|
| 912 |
+
"loss": 12.897845458984374,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.04508488095857393,
|
| 917 |
+
"grad_norm": 5.625839710235596,
|
| 918 |
+
"learning_rate": 0.0002999935604841711,
|
| 919 |
+
"loss": 13.06021728515625,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.04543168773517835,
|
| 924 |
+
"grad_norm": 7.4519877433776855,
|
| 925 |
+
"learning_rate": 0.00029999311187243257,
|
| 926 |
+
"loss": 12.539090728759765,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.04577849451178276,
|
| 931 |
+
"grad_norm": 5.8146071434021,
|
| 932 |
+
"learning_rate": 0.0002999926481563918,
|
| 933 |
+
"loss": 12.719709014892578,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.046125301288387176,
|
| 938 |
+
"grad_norm": 6.683107376098633,
|
| 939 |
+
"learning_rate": 0.0002999921693360955,
|
| 940 |
+
"loss": 13.211663818359375,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.04647210806499159,
|
| 945 |
+
"grad_norm": 5.467896938323975,
|
| 946 |
+
"learning_rate": 0.00029999167541159183,
|
| 947 |
+
"loss": 11.408037567138672,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.046818914841596004,
|
| 952 |
+
"grad_norm": 9.282605171203613,
|
| 953 |
+
"learning_rate": 0.00029999116638293054,
|
| 954 |
+
"loss": 12.204759979248047,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.04716572161820042,
|
| 959 |
+
"grad_norm": 6.287093162536621,
|
| 960 |
+
"learning_rate": 0.00029999064225016297,
|
| 961 |
+
"loss": 12.018901062011718,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.04751252839480483,
|
| 966 |
+
"grad_norm": 5.694602012634277,
|
| 967 |
+
"learning_rate": 0.00029999010301334176,
|
| 968 |
+
"loss": 11.694412231445312,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.04785933517140925,
|
| 973 |
+
"grad_norm": 5.267282962799072,
|
| 974 |
+
"learning_rate": 0.00029998954867252133,
|
| 975 |
+
"loss": 11.589154052734376,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.04820614194801366,
|
| 980 |
+
"grad_norm": 6.381001949310303,
|
| 981 |
+
"learning_rate": 0.00029998897922775737,
|
| 982 |
+
"loss": 11.301937103271484,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.04855294872461808,
|
| 987 |
+
"grad_norm": 4.900434970855713,
|
| 988 |
+
"learning_rate": 0.0002999883946791074,
|
| 989 |
+
"loss": 11.892779541015624,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.0488997555012225,
|
| 994 |
+
"grad_norm": 6.233642578125,
|
| 995 |
+
"learning_rate": 0.00029998779502663013,
|
| 996 |
+
"loss": 11.379587554931641,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.04924656227782691,
|
| 1001 |
+
"grad_norm": 4.936466693878174,
|
| 1002 |
+
"learning_rate": 0.000299987180270386,
|
| 1003 |
+
"loss": 11.712295532226562,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.049593369054431326,
|
| 1008 |
+
"grad_norm": 4.68892240524292,
|
| 1009 |
+
"learning_rate": 0.000299986550410437,
|
| 1010 |
+
"loss": 11.267239379882813,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.04994017583103574,
|
| 1015 |
+
"grad_norm": 4.828918933868408,
|
| 1016 |
+
"learning_rate": 0.00029998590544684636,
|
| 1017 |
+
"loss": 11.923808288574218,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.050286982607640154,
|
| 1022 |
+
"grad_norm": 5.1969380378723145,
|
| 1023 |
+
"learning_rate": 0.0002999852453796792,
|
| 1024 |
+
"loss": 11.340673828125,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.050633789384244565,
|
| 1029 |
+
"grad_norm": 5.5531325340271,
|
| 1030 |
+
"learning_rate": 0.00029998457020900197,
|
| 1031 |
+
"loss": 11.629721069335938,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.05098059616084898,
|
| 1036 |
+
"grad_norm": 4.863478183746338,
|
| 1037 |
+
"learning_rate": 0.0002999838799348826,
|
| 1038 |
+
"loss": 10.981313323974609,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.0513274029374534,
|
| 1043 |
+
"grad_norm": 5.0770368576049805,
|
| 1044 |
+
"learning_rate": 0.0002999831745573906,
|
| 1045 |
+
"loss": 10.801561737060547,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.05167420971405781,
|
| 1050 |
+
"grad_norm": 4.6786088943481445,
|
| 1051 |
+
"learning_rate": 0.00029998245407659703,
|
| 1052 |
+
"loss": 11.578643798828125,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.05202101649066223,
|
| 1057 |
+
"grad_norm": 5.357450485229492,
|
| 1058 |
+
"learning_rate": 0.00029998171849257446,
|
| 1059 |
+
"loss": 11.369083404541016,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.05236782326726664,
|
| 1064 |
+
"grad_norm": 5.213536739349365,
|
| 1065 |
+
"learning_rate": 0.00029998096780539695,
|
| 1066 |
+
"loss": 11.568294525146484,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.05271463004387106,
|
| 1071 |
+
"grad_norm": 5.254374027252197,
|
| 1072 |
+
"learning_rate": 0.00029998020201514006,
|
| 1073 |
+
"loss": 10.894474029541016,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.05306143682047547,
|
| 1078 |
+
"grad_norm": 4.989107131958008,
|
| 1079 |
+
"learning_rate": 0.00029997942112188093,
|
| 1080 |
+
"loss": 11.209683990478515,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.05340824359707989,
|
| 1085 |
+
"grad_norm": 4.966851234436035,
|
| 1086 |
+
"learning_rate": 0.00029997862512569823,
|
| 1087 |
+
"loss": 10.634807586669922,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.053755050373684304,
|
| 1092 |
+
"grad_norm": 4.828151702880859,
|
| 1093 |
+
"learning_rate": 0.000299977814026672,
|
| 1094 |
+
"loss": 11.01966552734375,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.054101857150288715,
|
| 1099 |
+
"grad_norm": 4.8093953132629395,
|
| 1100 |
+
"learning_rate": 0.0002999769878248841,
|
| 1101 |
+
"loss": 10.188558959960938,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.05444866392689313,
|
| 1106 |
+
"grad_norm": 4.421850204467773,
|
| 1107 |
+
"learning_rate": 0.00029997614652041763,
|
| 1108 |
+
"loss": 10.327834320068359,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.054795470703497544,
|
| 1113 |
+
"grad_norm": 5.034534931182861,
|
| 1114 |
+
"learning_rate": 0.00029997529011335723,
|
| 1115 |
+
"loss": 10.527660369873047,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.05514227748010196,
|
| 1120 |
+
"grad_norm": 5.502425670623779,
|
| 1121 |
+
"learning_rate": 0.00029997441860378927,
|
| 1122 |
+
"loss": 10.757402038574218,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.05548908425670638,
|
| 1127 |
+
"grad_norm": 8.620652198791504,
|
| 1128 |
+
"learning_rate": 0.00029997353199180145,
|
| 1129 |
+
"loss": 10.318006896972657,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.05583589103331079,
|
| 1134 |
+
"grad_norm": 4.712747573852539,
|
| 1135 |
+
"learning_rate": 0.00029997263027748306,
|
| 1136 |
+
"loss": 10.284696197509765,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.05618269780991521,
|
| 1141 |
+
"grad_norm": 4.792940139770508,
|
| 1142 |
+
"learning_rate": 0.00029997171346092495,
|
| 1143 |
+
"loss": 9.87569808959961,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.05652950458651962,
|
| 1148 |
+
"grad_norm": 4.812749862670898,
|
| 1149 |
+
"learning_rate": 0.00029997078154221933,
|
| 1150 |
+
"loss": 10.548392486572265,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.05687631136312404,
|
| 1155 |
+
"grad_norm": 5.6575422286987305,
|
| 1156 |
+
"learning_rate": 0.0002999698345214601,
|
| 1157 |
+
"loss": 10.364289855957031,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.05722311813972845,
|
| 1162 |
+
"grad_norm": 4.242788314819336,
|
| 1163 |
+
"learning_rate": 0.00029996887239874267,
|
| 1164 |
+
"loss": 10.221109008789062,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.057569924916332865,
|
| 1169 |
+
"grad_norm": 4.478567600250244,
|
| 1170 |
+
"learning_rate": 0.0002999678951741639,
|
| 1171 |
+
"loss": 9.714788818359375,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.05791673169293728,
|
| 1176 |
+
"grad_norm": 5.814032077789307,
|
| 1177 |
+
"learning_rate": 0.0002999669028478222,
|
| 1178 |
+
"loss": 10.347315216064453,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.058263538469541694,
|
| 1183 |
+
"grad_norm": 4.448582649230957,
|
| 1184 |
+
"learning_rate": 0.00029996589541981747,
|
| 1185 |
+
"loss": 10.337659454345703,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.05861034524614611,
|
| 1190 |
+
"grad_norm": 4.868472576141357,
|
| 1191 |
+
"learning_rate": 0.00029996487289025113,
|
| 1192 |
+
"loss": 10.064467620849609,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.05895715202275052,
|
| 1197 |
+
"grad_norm": 4.490790843963623,
|
| 1198 |
+
"learning_rate": 0.0002999638352592263,
|
| 1199 |
+
"loss": 9.652999877929688,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.05930395879935494,
|
| 1204 |
+
"grad_norm": 4.921995639801025,
|
| 1205 |
+
"learning_rate": 0.0002999627825268473,
|
| 1206 |
+
"loss": 9.934603881835937,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.05965076557595935,
|
| 1211 |
+
"grad_norm": 4.262001037597656,
|
| 1212 |
+
"learning_rate": 0.0002999617146932202,
|
| 1213 |
+
"loss": 9.353959655761718,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.05999757235256377,
|
| 1218 |
+
"grad_norm": 4.506113052368164,
|
| 1219 |
+
"learning_rate": 0.0002999606317584525,
|
| 1220 |
+
"loss": 9.367766571044921,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.06034437912916819,
|
| 1225 |
+
"grad_norm": 4.591375827789307,
|
| 1226 |
+
"learning_rate": 0.00029995953372265335,
|
| 1227 |
+
"loss": 9.564098358154297,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.0606911859057726,
|
| 1232 |
+
"grad_norm": 4.821197509765625,
|
| 1233 |
+
"learning_rate": 0.00029995842058593323,
|
| 1234 |
+
"loss": 10.153470611572265,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.061037992682377015,
|
| 1239 |
+
"grad_norm": 4.2808308601379395,
|
| 1240 |
+
"learning_rate": 0.0002999572923484042,
|
| 1241 |
+
"loss": 9.919622802734375,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.061384799458981426,
|
| 1246 |
+
"grad_norm": 4.291468143463135,
|
| 1247 |
+
"learning_rate": 0.0002999561490101801,
|
| 1248 |
+
"loss": 9.118783569335937,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.061731606235585844,
|
| 1253 |
+
"grad_norm": 4.747788906097412,
|
| 1254 |
+
"learning_rate": 0.00029995499057137575,
|
| 1255 |
+
"loss": 9.513807678222657,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.06207841301219026,
|
| 1260 |
+
"grad_norm": 4.285642147064209,
|
| 1261 |
+
"learning_rate": 0.000299953817032108,
|
| 1262 |
+
"loss": 9.9878173828125,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.06242521978879467,
|
| 1267 |
+
"grad_norm": 4.473206520080566,
|
| 1268 |
+
"learning_rate": 0.00029995262839249497,
|
| 1269 |
+
"loss": 9.633001708984375,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.06277202656539908,
|
| 1274 |
+
"grad_norm": 4.4688005447387695,
|
| 1275 |
+
"learning_rate": 0.00029995142465265637,
|
| 1276 |
+
"loss": 9.59437255859375,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.06311883334200351,
|
| 1281 |
+
"grad_norm": 4.453664779663086,
|
| 1282 |
+
"learning_rate": 0.0002999502058127134,
|
| 1283 |
+
"loss": 8.898344421386719,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.06346564011860792,
|
| 1288 |
+
"grad_norm": 4.2538275718688965,
|
| 1289 |
+
"learning_rate": 0.0002999489718727888,
|
| 1290 |
+
"loss": 9.250914764404296,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.06381244689521233,
|
| 1295 |
+
"grad_norm": 4.033658981323242,
|
| 1296 |
+
"learning_rate": 0.0002999477228330069,
|
| 1297 |
+
"loss": 9.316190338134765,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.06415925367181674,
|
| 1302 |
+
"grad_norm": 4.747572898864746,
|
| 1303 |
+
"learning_rate": 0.00029994645869349334,
|
| 1304 |
+
"loss": 8.979618072509766,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.06450606044842117,
|
| 1309 |
+
"grad_norm": 3.9432504177093506,
|
| 1310 |
+
"learning_rate": 0.0002999451794543755,
|
| 1311 |
+
"loss": 9.21793441772461,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.06485286722502558,
|
| 1316 |
+
"grad_norm": 4.528207302093506,
|
| 1317 |
+
"learning_rate": 0.0002999438851157822,
|
| 1318 |
+
"loss": 9.081039428710938,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.06519967400162999,
|
| 1323 |
+
"grad_norm": 4.3286542892456055,
|
| 1324 |
+
"learning_rate": 0.0002999425756778438,
|
| 1325 |
+
"loss": 9.034320831298828,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.06554648077823441,
|
| 1330 |
+
"grad_norm": 5.157027721405029,
|
| 1331 |
+
"learning_rate": 0.0002999412511406921,
|
| 1332 |
+
"loss": 8.857444763183594,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.06589328755483882,
|
| 1337 |
+
"grad_norm": 4.572315692901611,
|
| 1338 |
+
"learning_rate": 0.00029993991150446055,
|
| 1339 |
+
"loss": 9.055838012695313,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.06624009433144323,
|
| 1344 |
+
"grad_norm": 4.572640419006348,
|
| 1345 |
+
"learning_rate": 0.00029993855676928397,
|
| 1346 |
+
"loss": 9.332460021972656,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.06658690110804766,
|
| 1351 |
+
"grad_norm": 5.196224212646484,
|
| 1352 |
+
"learning_rate": 0.00029993718693529885,
|
| 1353 |
+
"loss": 8.875639343261719,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.06693370788465207,
|
| 1358 |
+
"grad_norm": 4.320059299468994,
|
| 1359 |
+
"learning_rate": 0.0002999358020026431,
|
| 1360 |
+
"loss": 8.493698120117188,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.06728051466125648,
|
| 1365 |
+
"grad_norm": 5.197443008422852,
|
| 1366 |
+
"learning_rate": 0.0002999344019714562,
|
| 1367 |
+
"loss": 8.634622192382812,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.06762732143786089,
|
| 1372 |
+
"grad_norm": 5.580541133880615,
|
| 1373 |
+
"learning_rate": 0.0002999329868418791,
|
| 1374 |
+
"loss": 9.131027984619141,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.06797412821446532,
|
| 1379 |
+
"grad_norm": 4.569746017456055,
|
| 1380 |
+
"learning_rate": 0.00029993155661405435,
|
| 1381 |
+
"loss": 8.806238555908203,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.06832093499106973,
|
| 1386 |
+
"grad_norm": 5.034822940826416,
|
| 1387 |
+
"learning_rate": 0.00029993011128812597,
|
| 1388 |
+
"loss": 8.735447692871094,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.06866774176767414,
|
| 1393 |
+
"grad_norm": 5.717297077178955,
|
| 1394 |
+
"learning_rate": 0.00029992865086423944,
|
| 1395 |
+
"loss": 8.358311462402344,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.06901454854427856,
|
| 1400 |
+
"grad_norm": 5.722506046295166,
|
| 1401 |
+
"learning_rate": 0.00029992717534254194,
|
| 1402 |
+
"loss": 8.968340301513672,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.06936135532088297,
|
| 1407 |
+
"grad_norm": 4.75577974319458,
|
| 1408 |
+
"learning_rate": 0.0002999256847231819,
|
| 1409 |
+
"loss": 8.24468765258789,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.06970816209748738,
|
| 1414 |
+
"grad_norm": 4.7462029457092285,
|
| 1415 |
+
"learning_rate": 0.00029992417900630953,
|
| 1416 |
+
"loss": 9.001568603515626,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.0700549688740918,
|
| 1421 |
+
"grad_norm": 4.48958158493042,
|
| 1422 |
+
"learning_rate": 0.00029992265819207646,
|
| 1423 |
+
"loss": 9.029542541503906,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.07040177565069622,
|
| 1428 |
+
"grad_norm": 4.929230213165283,
|
| 1429 |
+
"learning_rate": 0.0002999211222806358,
|
| 1430 |
+
"loss": 8.544095611572265,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.07074858242730063,
|
| 1435 |
+
"grad_norm": 5.174428939819336,
|
| 1436 |
+
"learning_rate": 0.0002999195712721422,
|
| 1437 |
+
"loss": 9.152230834960937,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.07109538920390504,
|
| 1442 |
+
"grad_norm": 5.165338516235352,
|
| 1443 |
+
"learning_rate": 0.00029991800516675194,
|
| 1444 |
+
"loss": 8.47047119140625,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.07144219598050947,
|
| 1449 |
+
"grad_norm": 4.739821910858154,
|
| 1450 |
+
"learning_rate": 0.00029991642396462264,
|
| 1451 |
+
"loss": 8.763507080078124,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.07178900275711388,
|
| 1456 |
+
"grad_norm": 5.1035637855529785,
|
| 1457 |
+
"learning_rate": 0.00029991482766591354,
|
| 1458 |
+
"loss": 8.246512603759765,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.07213580953371829,
|
| 1463 |
+
"grad_norm": 5.471373558044434,
|
| 1464 |
+
"learning_rate": 0.0002999132162707854,
|
| 1465 |
+
"loss": 8.864594268798829,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.0724826163103227,
|
| 1470 |
+
"grad_norm": 5.771513938903809,
|
| 1471 |
+
"learning_rate": 0.0002999115897794005,
|
| 1472 |
+
"loss": 8.889707946777344,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.07282942308692712,
|
| 1477 |
+
"grad_norm": 5.27165412902832,
|
| 1478 |
+
"learning_rate": 0.00029990994819192255,
|
| 1479 |
+
"loss": 8.567472839355469,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.07317622986353153,
|
| 1484 |
+
"grad_norm": 4.975783824920654,
|
| 1485 |
+
"learning_rate": 0.00029990829150851697,
|
| 1486 |
+
"loss": 8.457414245605468,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.07352303664013594,
|
| 1491 |
+
"grad_norm": 6.3124680519104,
|
| 1492 |
+
"learning_rate": 0.00029990661972935053,
|
| 1493 |
+
"loss": 8.194977569580079,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.07386984341674037,
|
| 1498 |
+
"grad_norm": 4.776012897491455,
|
| 1499 |
+
"learning_rate": 0.00029990493285459157,
|
| 1500 |
+
"loss": 8.464956665039063,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.07421665019334478,
|
| 1505 |
+
"grad_norm": 4.533716678619385,
|
| 1506 |
+
"learning_rate": 0.00029990323088440997,
|
| 1507 |
+
"loss": 8.268666839599609,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.07456345696994919,
|
| 1512 |
+
"grad_norm": 5.62864875793457,
|
| 1513 |
+
"learning_rate": 0.00029990151381897715,
|
| 1514 |
+
"loss": 7.639604949951172,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.0749102637465536,
|
| 1519 |
+
"grad_norm": 4.756488800048828,
|
| 1520 |
+
"learning_rate": 0.000299899781658466,
|
| 1521 |
+
"loss": 8.60247573852539,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.07525707052315803,
|
| 1526 |
+
"grad_norm": 5.337765693664551,
|
| 1527 |
+
"learning_rate": 0.00029989803440305093,
|
| 1528 |
+
"loss": 8.40919189453125,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.07560387729976244,
|
| 1533 |
+
"grad_norm": 4.989221572875977,
|
| 1534 |
+
"learning_rate": 0.0002998962720529079,
|
| 1535 |
+
"loss": 8.433936309814452,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.07595068407636685,
|
| 1540 |
+
"grad_norm": 5.895046710968018,
|
| 1541 |
+
"learning_rate": 0.00029989449460821434,
|
| 1542 |
+
"loss": 8.606028747558593,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.07629749085297127,
|
| 1547 |
+
"grad_norm": 4.604063034057617,
|
| 1548 |
+
"learning_rate": 0.0002998927020691493,
|
| 1549 |
+
"loss": 8.863967132568359,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.07664429762957568,
|
| 1554 |
+
"grad_norm": 4.635347366333008,
|
| 1555 |
+
"learning_rate": 0.0002998908944358932,
|
| 1556 |
+
"loss": 8.632648468017578,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.0769911044061801,
|
| 1561 |
+
"grad_norm": 5.013646125793457,
|
| 1562 |
+
"learning_rate": 0.0002998890717086282,
|
| 1563 |
+
"loss": 8.241293334960938,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.0773379111827845,
|
| 1568 |
+
"grad_norm": 5.209301948547363,
|
| 1569 |
+
"learning_rate": 0.0002998872338875378,
|
| 1570 |
+
"loss": 8.100372314453125,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.07768471795938893,
|
| 1575 |
+
"grad_norm": 5.827475547790527,
|
| 1576 |
+
"learning_rate": 0.00029988538097280696,
|
| 1577 |
+
"loss": 7.576439666748047,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.07803152473599334,
|
| 1582 |
+
"grad_norm": 5.064399242401123,
|
| 1583 |
+
"learning_rate": 0.00029988351296462247,
|
| 1584 |
+
"loss": 8.33707275390625,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.07837833151259775,
|
| 1589 |
+
"grad_norm": 5.7265706062316895,
|
| 1590 |
+
"learning_rate": 0.0002998816298631723,
|
| 1591 |
+
"loss": 8.343417358398437,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.07872513828920218,
|
| 1596 |
+
"grad_norm": 5.61683988571167,
|
| 1597 |
+
"learning_rate": 0.00029987973166864605,
|
| 1598 |
+
"loss": 7.654642486572266,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.07907194506580659,
|
| 1603 |
+
"grad_norm": 6.139464855194092,
|
| 1604 |
+
"learning_rate": 0.0002998778183812349,
|
| 1605 |
+
"loss": 8.174369812011719,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.079418751842411,
|
| 1610 |
+
"grad_norm": 5.3133225440979,
|
| 1611 |
+
"learning_rate": 0.0002998758900011316,
|
| 1612 |
+
"loss": 8.249425506591797,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.07976555861901541,
|
| 1617 |
+
"grad_norm": 5.132144927978516,
|
| 1618 |
+
"learning_rate": 0.0002998739465285303,
|
| 1619 |
+
"loss": 8.233519744873046,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.08011236539561983,
|
| 1624 |
+
"grad_norm": 6.415826320648193,
|
| 1625 |
+
"learning_rate": 0.00029987198796362666,
|
| 1626 |
+
"loss": 8.221486663818359,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.08045917217222424,
|
| 1631 |
+
"grad_norm": 5.217477798461914,
|
| 1632 |
+
"learning_rate": 0.0002998700143066179,
|
| 1633 |
+
"loss": 8.524445343017579,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.08080597894882866,
|
| 1638 |
+
"grad_norm": 5.741121292114258,
|
| 1639 |
+
"learning_rate": 0.0002998680255577028,
|
| 1640 |
+
"loss": 8.453343200683594,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.08115278572543308,
|
| 1645 |
+
"grad_norm": 6.204251289367676,
|
| 1646 |
+
"learning_rate": 0.0002998660217170816,
|
| 1647 |
+
"loss": 8.372715759277344,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.08149959250203749,
|
| 1652 |
+
"grad_norm": 7.696920394897461,
|
| 1653 |
+
"learning_rate": 0.00029986400278495624,
|
| 1654 |
+
"loss": 8.396324157714844,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.0818463992786419,
|
| 1659 |
+
"grad_norm": 6.234743118286133,
|
| 1660 |
+
"learning_rate": 0.00029986196876152977,
|
| 1661 |
+
"loss": 8.096724700927734,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.08219320605524633,
|
| 1666 |
+
"grad_norm": 4.549152374267578,
|
| 1667 |
+
"learning_rate": 0.0002998599196470072,
|
| 1668 |
+
"loss": 7.865915679931641,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.08254001283185074,
|
| 1673 |
+
"grad_norm": 5.453365325927734,
|
| 1674 |
+
"learning_rate": 0.00029985785544159484,
|
| 1675 |
+
"loss": 7.964163970947266,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.08288681960845515,
|
| 1680 |
+
"grad_norm": 4.967399597167969,
|
| 1681 |
+
"learning_rate": 0.0002998557761455005,
|
| 1682 |
+
"loss": 7.947079467773437,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.08323362638505956,
|
| 1687 |
+
"grad_norm": 6.34585428237915,
|
| 1688 |
+
"learning_rate": 0.00029985368175893364,
|
| 1689 |
+
"loss": 8.035052490234374,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.08358043316166398,
|
| 1694 |
+
"grad_norm": 5.457056999206543,
|
| 1695 |
+
"learning_rate": 0.0002998515722821051,
|
| 1696 |
+
"loss": 7.74532470703125,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.0839272399382684,
|
| 1701 |
+
"grad_norm": 4.875389099121094,
|
| 1702 |
+
"learning_rate": 0.00029984944771522734,
|
| 1703 |
+
"loss": 7.901406860351562,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.0842740467148728,
|
| 1708 |
+
"grad_norm": 5.598247051239014,
|
| 1709 |
+
"learning_rate": 0.00029984730805851425,
|
| 1710 |
+
"loss": 8.011869812011719,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.08462085349147723,
|
| 1715 |
+
"grad_norm": 5.847703456878662,
|
| 1716 |
+
"learning_rate": 0.0002998451533121814,
|
| 1717 |
+
"loss": 8.203024291992188,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.08496766026808164,
|
| 1722 |
+
"grad_norm": 7.487412452697754,
|
| 1723 |
+
"learning_rate": 0.00029984298347644566,
|
| 1724 |
+
"loss": 7.6920112609863285,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.08531446704468605,
|
| 1729 |
+
"grad_norm": 6.7232255935668945,
|
| 1730 |
+
"learning_rate": 0.0002998407985515257,
|
| 1731 |
+
"loss": 7.558551788330078,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.08566127382129046,
|
| 1736 |
+
"grad_norm": 7.134231090545654,
|
| 1737 |
+
"learning_rate": 0.0002998385985376413,
|
| 1738 |
+
"loss": 7.462080383300782,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.08600808059789489,
|
| 1743 |
+
"grad_norm": 5.4149909019470215,
|
| 1744 |
+
"learning_rate": 0.00029983638343501415,
|
| 1745 |
+
"loss": 8.575581359863282,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.0863548873744993,
|
| 1750 |
+
"grad_norm": 6.803240776062012,
|
| 1751 |
+
"learning_rate": 0.0002998341532438674,
|
| 1752 |
+
"loss": 7.64642333984375,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.08670169415110371,
|
| 1757 |
+
"grad_norm": 5.699243068695068,
|
| 1758 |
+
"learning_rate": 0.0002998319079644254,
|
| 1759 |
+
"loss": 7.960725402832031,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.08704850092770813,
|
| 1764 |
+
"grad_norm": 6.425755500793457,
|
| 1765 |
+
"learning_rate": 0.00029982964759691435,
|
| 1766 |
+
"loss": 7.6616157531738285,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.08739530770431254,
|
| 1771 |
+
"grad_norm": 5.376076698303223,
|
| 1772 |
+
"learning_rate": 0.00029982737214156194,
|
| 1773 |
+
"loss": 7.7346656799316404,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.08774211448091696,
|
| 1778 |
+
"grad_norm": 6.2727484703063965,
|
| 1779 |
+
"learning_rate": 0.0002998250815985972,
|
| 1780 |
+
"loss": 7.836188507080078,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.08808892125752137,
|
| 1785 |
+
"grad_norm": 5.631091117858887,
|
| 1786 |
+
"learning_rate": 0.0002998227759682509,
|
| 1787 |
+
"loss": 7.705145263671875,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.08843572803412579,
|
| 1792 |
+
"grad_norm": 6.217864036560059,
|
| 1793 |
+
"learning_rate": 0.00029982045525075513,
|
| 1794 |
+
"loss": 8.142205047607423,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.0887825348107302,
|
| 1799 |
+
"grad_norm": 5.994324684143066,
|
| 1800 |
+
"learning_rate": 0.00029981811944634364,
|
| 1801 |
+
"loss": 7.400997924804687,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.08912934158733461,
|
| 1806 |
+
"grad_norm": 4.685929298400879,
|
| 1807 |
+
"learning_rate": 0.00029981576855525165,
|
| 1808 |
+
"loss": 8.070906829833984,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.08947614836393904,
|
| 1813 |
+
"grad_norm": 5.380947589874268,
|
| 1814 |
+
"learning_rate": 0.00029981340257771586,
|
| 1815 |
+
"loss": 7.440650177001953,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.08982295514054345,
|
| 1820 |
+
"grad_norm": 5.548213005065918,
|
| 1821 |
+
"learning_rate": 0.00029981102151397445,
|
| 1822 |
+
"loss": 7.631075286865235,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.09016976191714786,
|
| 1827 |
+
"grad_norm": 7.113351821899414,
|
| 1828 |
+
"learning_rate": 0.00029980862536426736,
|
| 1829 |
+
"loss": 7.411952972412109,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.09051656869375227,
|
| 1834 |
+
"grad_norm": 4.830175399780273,
|
| 1835 |
+
"learning_rate": 0.0002998062141288357,
|
| 1836 |
+
"loss": 8.161653900146485,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.0908633754703567,
|
| 1841 |
+
"grad_norm": 5.5586018562316895,
|
| 1842 |
+
"learning_rate": 0.00029980378780792243,
|
| 1843 |
+
"loss": 7.329508209228516,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.0912101822469611,
|
| 1848 |
+
"grad_norm": 6.112547397613525,
|
| 1849 |
+
"learning_rate": 0.0002998013464017719,
|
| 1850 |
+
"loss": 7.714527893066406,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.09155698902356552,
|
| 1855 |
+
"grad_norm": 4.658982276916504,
|
| 1856 |
+
"learning_rate": 0.00029979888991062977,
|
| 1857 |
+
"loss": 7.437145233154297,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.09190379580016994,
|
| 1862 |
+
"grad_norm": 6.236598014831543,
|
| 1863 |
+
"learning_rate": 0.0002997964183347436,
|
| 1864 |
+
"loss": 7.697496795654297,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.09225060257677435,
|
| 1869 |
+
"grad_norm": 4.987429618835449,
|
| 1870 |
+
"learning_rate": 0.00029979393167436214,
|
| 1871 |
+
"loss": 7.137907409667969,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.09259740935337876,
|
| 1876 |
+
"grad_norm": 7.033418655395508,
|
| 1877 |
+
"learning_rate": 0.0002997914299297359,
|
| 1878 |
+
"loss": 7.323359680175781,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.09294421612998317,
|
| 1883 |
+
"grad_norm": 5.532907009124756,
|
| 1884 |
+
"learning_rate": 0.00029978891310111674,
|
| 1885 |
+
"loss": 7.454933166503906,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.0932910229065876,
|
| 1890 |
+
"grad_norm": 6.004257678985596,
|
| 1891 |
+
"learning_rate": 0.00029978638118875815,
|
| 1892 |
+
"loss": 7.570569610595703,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.09363782968319201,
|
| 1897 |
+
"grad_norm": 6.463276386260986,
|
| 1898 |
+
"learning_rate": 0.0002997838341929151,
|
| 1899 |
+
"loss": 7.730841827392578,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.09398463645979642,
|
| 1904 |
+
"grad_norm": 6.122608184814453,
|
| 1905 |
+
"learning_rate": 0.000299781272113844,
|
| 1906 |
+
"loss": 7.185985565185547,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.09433144323640084,
|
| 1911 |
+
"grad_norm": 5.712291717529297,
|
| 1912 |
+
"learning_rate": 0.00029977869495180287,
|
| 1913 |
+
"loss": 8.223866271972657,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.09467825001300526,
|
| 1918 |
+
"grad_norm": 6.126311302185059,
|
| 1919 |
+
"learning_rate": 0.00029977610270705133,
|
| 1920 |
+
"loss": 7.804103088378906,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.09502505678960967,
|
| 1925 |
+
"grad_norm": 4.903704643249512,
|
| 1926 |
+
"learning_rate": 0.0002997734953798503,
|
| 1927 |
+
"loss": 7.276256561279297,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.09537186356621409,
|
| 1932 |
+
"grad_norm": 6.626465797424316,
|
| 1933 |
+
"learning_rate": 0.0002997708729704624,
|
| 1934 |
+
"loss": 7.943316650390625,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.0957186703428185,
|
| 1939 |
+
"grad_norm": 6.401085376739502,
|
| 1940 |
+
"learning_rate": 0.0002997682354791517,
|
| 1941 |
+
"loss": 7.571746826171875,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.09606547711942291,
|
| 1946 |
+
"grad_norm": 6.731732368469238,
|
| 1947 |
+
"learning_rate": 0.00029976558290618373,
|
| 1948 |
+
"loss": 8.095571136474609,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.09641228389602732,
|
| 1953 |
+
"grad_norm": 6.18290901184082,
|
| 1954 |
+
"learning_rate": 0.00029976291525182574,
|
| 1955 |
+
"loss": 7.459332275390625,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.09675909067263175,
|
| 1960 |
+
"grad_norm": 4.505118370056152,
|
| 1961 |
+
"learning_rate": 0.00029976023251634623,
|
| 1962 |
+
"loss": 8.029454040527344,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.09710589744923616,
|
| 1967 |
+
"grad_norm": 7.047544479370117,
|
| 1968 |
+
"learning_rate": 0.00029975753470001543,
|
| 1969 |
+
"loss": 7.250439453125,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.09745270422584057,
|
| 1974 |
+
"grad_norm": 7.906260967254639,
|
| 1975 |
+
"learning_rate": 0.00029975482180310495,
|
| 1976 |
+
"loss": 7.352965545654297,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.097799511002445,
|
| 1981 |
+
"grad_norm": 5.261754989624023,
|
| 1982 |
+
"learning_rate": 0.00029975209382588806,
|
| 1983 |
+
"loss": 7.319126892089844,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.0981463177790494,
|
| 1988 |
+
"grad_norm": 5.247056007385254,
|
| 1989 |
+
"learning_rate": 0.0002997493507686394,
|
| 1990 |
+
"loss": 7.372547912597656,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.09849312455565382,
|
| 1995 |
+
"grad_norm": 5.888732433319092,
|
| 1996 |
+
"learning_rate": 0.00029974659263163516,
|
| 1997 |
+
"loss": 7.207266235351563,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.09883993133225823,
|
| 2002 |
+
"grad_norm": 6.898653030395508,
|
| 2003 |
+
"learning_rate": 0.0002997438194151532,
|
| 2004 |
+
"loss": 7.474554443359375,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.09918673810886265,
|
| 2009 |
+
"grad_norm": 5.344205379486084,
|
| 2010 |
+
"learning_rate": 0.0002997410311194727,
|
| 2011 |
+
"loss": 7.531422424316406,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.09953354488546706,
|
| 2016 |
+
"grad_norm": 8.75732421875,
|
| 2017 |
+
"learning_rate": 0.00029973822774487447,
|
| 2018 |
+
"loss": 7.705049896240235,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.09988035166207147,
|
| 2023 |
+
"grad_norm": 5.712015151977539,
|
| 2024 |
+
"learning_rate": 0.0002997354092916408,
|
| 2025 |
+
"loss": 7.514460754394531,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.1002271584386759,
|
| 2030 |
+
"grad_norm": 6.071193218231201,
|
| 2031 |
+
"learning_rate": 0.0002997325757600555,
|
| 2032 |
+
"loss": 8.2220947265625,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.10057396521528031,
|
| 2037 |
+
"grad_norm": 4.527872085571289,
|
| 2038 |
+
"learning_rate": 0.00029972972715040395,
|
| 2039 |
+
"loss": 7.812737274169922,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 0.10092077199188472,
|
| 2044 |
+
"grad_norm": 5.524644374847412,
|
| 2045 |
+
"learning_rate": 0.00029972686346297294,
|
| 2046 |
+
"loss": 7.216932678222657,
|
| 2047 |
+
"step": 2910
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 0.10126757876848913,
|
| 2051 |
+
"grad_norm": 5.723361015319824,
|
| 2052 |
+
"learning_rate": 0.0002997239846980509,
|
| 2053 |
+
"loss": 7.164757537841797,
|
| 2054 |
+
"step": 2920
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 0.10161438554509356,
|
| 2058 |
+
"grad_norm": 6.528432369232178,
|
| 2059 |
+
"learning_rate": 0.0002997210908559277,
|
| 2060 |
+
"loss": 7.451429748535157,
|
| 2061 |
+
"step": 2930
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 0.10196119232169797,
|
| 2065 |
+
"grad_norm": 6.940042018890381,
|
| 2066 |
+
"learning_rate": 0.00029971818193689473,
|
| 2067 |
+
"loss": 7.0673583984375,
|
| 2068 |
+
"step": 2940
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.10230799909830238,
|
| 2072 |
+
"grad_norm": 5.519460678100586,
|
| 2073 |
+
"learning_rate": 0.00029971525794124495,
|
| 2074 |
+
"loss": 7.4651947021484375,
|
| 2075 |
+
"step": 2950
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 0.1026548058749068,
|
| 2079 |
+
"grad_norm": 5.205141544342041,
|
| 2080 |
+
"learning_rate": 0.0002997123188692728,
|
| 2081 |
+
"loss": 7.1986549377441404,
|
| 2082 |
+
"step": 2960
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.10300161265151121,
|
| 2086 |
+
"grad_norm": 4.912453651428223,
|
| 2087 |
+
"learning_rate": 0.0002997093647212742,
|
| 2088 |
+
"loss": 7.043345642089844,
|
| 2089 |
+
"step": 2970
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 0.10334841942811562,
|
| 2093 |
+
"grad_norm": 5.264575481414795,
|
| 2094 |
+
"learning_rate": 0.00029970639549754667,
|
| 2095 |
+
"loss": 7.150616455078125,
|
| 2096 |
+
"step": 2980
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 0.10369522620472003,
|
| 2100 |
+
"grad_norm": 8.516182899475098,
|
| 2101 |
+
"learning_rate": 0.0002997034111983892,
|
| 2102 |
+
"loss": 6.536961364746094,
|
| 2103 |
+
"step": 2990
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 0.10404203298132446,
|
| 2107 |
+
"grad_norm": 5.379635810852051,
|
| 2108 |
+
"learning_rate": 0.00029970041182410235,
|
| 2109 |
+
"loss": 7.469226837158203,
|
| 2110 |
+
"step": 3000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.10438883975792887,
|
| 2114 |
+
"grad_norm": 4.950097560882568,
|
| 2115 |
+
"learning_rate": 0.0002996973973749882,
|
| 2116 |
+
"loss": 7.181357574462891,
|
| 2117 |
+
"step": 3010
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.10473564653453328,
|
| 2121 |
+
"grad_norm": 4.616489410400391,
|
| 2122 |
+
"learning_rate": 0.00029969436785135006,
|
| 2123 |
+
"loss": 7.631926727294922,
|
| 2124 |
+
"step": 3020
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.1050824533111377,
|
| 2128 |
+
"grad_norm": 5.835964202880859,
|
| 2129 |
+
"learning_rate": 0.0002996913232534933,
|
| 2130 |
+
"loss": 7.360746002197265,
|
| 2131 |
+
"step": 3030
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.10542926008774212,
|
| 2135 |
+
"grad_norm": 4.751529693603516,
|
| 2136 |
+
"learning_rate": 0.00029968826358172437,
|
| 2137 |
+
"loss": 7.583438110351563,
|
| 2138 |
+
"step": 3040
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.10577606686434653,
|
| 2142 |
+
"grad_norm": 5.94398307800293,
|
| 2143 |
+
"learning_rate": 0.00029968518883635134,
|
| 2144 |
+
"loss": 7.026883697509765,
|
| 2145 |
+
"step": 3050
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.10612287364095094,
|
| 2149 |
+
"grad_norm": 7.210505485534668,
|
| 2150 |
+
"learning_rate": 0.0002996820990176839,
|
| 2151 |
+
"loss": 7.462823486328125,
|
| 2152 |
+
"step": 3060
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.10646968041755536,
|
| 2156 |
+
"grad_norm": 5.726904392242432,
|
| 2157 |
+
"learning_rate": 0.00029967899412603323,
|
| 2158 |
+
"loss": 7.295240020751953,
|
| 2159 |
+
"step": 3070
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.10681648719415977,
|
| 2163 |
+
"grad_norm": 5.8899383544921875,
|
| 2164 |
+
"learning_rate": 0.00029967587416171194,
|
| 2165 |
+
"loss": 7.170245361328125,
|
| 2166 |
+
"step": 3080
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.10716329397076418,
|
| 2170 |
+
"grad_norm": 5.578789710998535,
|
| 2171 |
+
"learning_rate": 0.0002996727391250342,
|
| 2172 |
+
"loss": 7.342673492431641,
|
| 2173 |
+
"step": 3090
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.10751010074736861,
|
| 2177 |
+
"grad_norm": 7.131217956542969,
|
| 2178 |
+
"learning_rate": 0.0002996695890163157,
|
| 2179 |
+
"loss": 7.228727722167969,
|
| 2180 |
+
"step": 3100
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.10785690752397302,
|
| 2184 |
+
"grad_norm": 4.9802775382995605,
|
| 2185 |
+
"learning_rate": 0.0002996664238358737,
|
| 2186 |
+
"loss": 7.686077880859375,
|
| 2187 |
+
"step": 3110
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.10820371430057743,
|
| 2191 |
+
"grad_norm": 7.20836877822876,
|
| 2192 |
+
"learning_rate": 0.000299663243584027,
|
| 2193 |
+
"loss": 6.775941467285156,
|
| 2194 |
+
"step": 3120
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.10855052107718184,
|
| 2198 |
+
"grad_norm": 6.248165607452393,
|
| 2199 |
+
"learning_rate": 0.0002996600482610957,
|
| 2200 |
+
"loss": 7.430239868164063,
|
| 2201 |
+
"step": 3130
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.10889732785378627,
|
| 2205 |
+
"grad_norm": 4.04344367980957,
|
| 2206 |
+
"learning_rate": 0.0002996568378674016,
|
| 2207 |
+
"loss": 6.764356231689453,
|
| 2208 |
+
"step": 3140
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.10924413463039068,
|
| 2212 |
+
"grad_norm": 5.1497907638549805,
|
| 2213 |
+
"learning_rate": 0.000299653612403268,
|
| 2214 |
+
"loss": 7.198393249511719,
|
| 2215 |
+
"step": 3150
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.10959094140699509,
|
| 2219 |
+
"grad_norm": 4.548824310302734,
|
| 2220 |
+
"learning_rate": 0.0002996503718690198,
|
| 2221 |
+
"loss": 7.365312957763672,
|
| 2222 |
+
"step": 3160
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.10993774818359951,
|
| 2226 |
+
"grad_norm": 5.565645694732666,
|
| 2227 |
+
"learning_rate": 0.0002996471162649832,
|
| 2228 |
+
"loss": 7.002183532714843,
|
| 2229 |
+
"step": 3170
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.11028455496020392,
|
| 2233 |
+
"grad_norm": 5.415268898010254,
|
| 2234 |
+
"learning_rate": 0.00029964384559148615,
|
| 2235 |
+
"loss": 7.195500183105469,
|
| 2236 |
+
"step": 3180
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.11063136173680833,
|
| 2240 |
+
"grad_norm": 5.401646614074707,
|
| 2241 |
+
"learning_rate": 0.00029964055984885786,
|
| 2242 |
+
"loss": 7.308545684814453,
|
| 2243 |
+
"step": 3190
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.11097816851341276,
|
| 2247 |
+
"grad_norm": 36.28436279296875,
|
| 2248 |
+
"learning_rate": 0.0002996379204052031,
|
| 2249 |
+
"loss": 7.051918029785156,
|
| 2250 |
+
"step": 3200
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.11132497529001717,
|
| 2254 |
+
"grad_norm": 75.87548828125,
|
| 2255 |
+
"learning_rate": 0.0002996346075389736,
|
| 2256 |
+
"loss": 7.859667205810547,
|
| 2257 |
+
"step": 3210
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.11167178206662158,
|
| 2261 |
+
"grad_norm": 6.471720218658447,
|
| 2262 |
+
"learning_rate": 0.0002996312796045432,
|
| 2263 |
+
"loss": 6.975923919677735,
|
| 2264 |
+
"step": 3220
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.11201858884322599,
|
| 2268 |
+
"grad_norm": 8.988420486450195,
|
| 2269 |
+
"learning_rate": 0.00029962793660224705,
|
| 2270 |
+
"loss": 7.185982513427734,
|
| 2271 |
+
"step": 3230
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.11236539561983042,
|
| 2275 |
+
"grad_norm": 7.521701335906982,
|
| 2276 |
+
"learning_rate": 0.0002996245785324218,
|
| 2277 |
+
"loss": 7.528837585449219,
|
| 2278 |
+
"step": 3240
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.11271220239643483,
|
| 2282 |
+
"grad_norm": 12.23703384399414,
|
| 2283 |
+
"learning_rate": 0.0002996212053954055,
|
| 2284 |
+
"loss": 7.36700439453125,
|
| 2285 |
+
"step": 3250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.11305900917303924,
|
| 2289 |
+
"grad_norm": 6.668253421783447,
|
| 2290 |
+
"learning_rate": 0.000299617817191538,
|
| 2291 |
+
"loss": 6.847956848144531,
|
| 2292 |
+
"step": 3260
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.11340581594964366,
|
| 2296 |
+
"grad_norm": 6.101006507873535,
|
| 2297 |
+
"learning_rate": 0.0002996144139211604,
|
| 2298 |
+
"loss": 6.717323303222656,
|
| 2299 |
+
"step": 3270
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.11375262272624807,
|
| 2303 |
+
"grad_norm": 6.196361541748047,
|
| 2304 |
+
"learning_rate": 0.00029961099558461536,
|
| 2305 |
+
"loss": 7.770551300048828,
|
| 2306 |
+
"step": 3280
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.11409942950285248,
|
| 2310 |
+
"grad_norm": 4.512564182281494,
|
| 2311 |
+
"learning_rate": 0.00029960756218224717,
|
| 2312 |
+
"loss": 6.8745979309082035,
|
| 2313 |
+
"step": 3290
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.1144462362794569,
|
| 2317 |
+
"grad_norm": 5.664703845977783,
|
| 2318 |
+
"learning_rate": 0.0002996041137144016,
|
| 2319 |
+
"loss": 7.512083435058594,
|
| 2320 |
+
"step": 3300
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.11479304305606132,
|
| 2324 |
+
"grad_norm": 5.180612087249756,
|
| 2325 |
+
"learning_rate": 0.0002996006501814258,
|
| 2326 |
+
"loss": 7.012033843994141,
|
| 2327 |
+
"step": 3310
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.11513984983266573,
|
| 2331 |
+
"grad_norm": 5.868711471557617,
|
| 2332 |
+
"learning_rate": 0.0002995971715836687,
|
| 2333 |
+
"loss": 6.600114440917968,
|
| 2334 |
+
"step": 3320
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.11548665660927014,
|
| 2338 |
+
"grad_norm": 6.995846271514893,
|
| 2339 |
+
"learning_rate": 0.00029959367792148047,
|
| 2340 |
+
"loss": 7.09898681640625,
|
| 2341 |
+
"step": 3330
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.11583346338587457,
|
| 2345 |
+
"grad_norm": 5.324821472167969,
|
| 2346 |
+
"learning_rate": 0.000299590169195213,
|
| 2347 |
+
"loss": 7.294175720214843,
|
| 2348 |
+
"step": 3340
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.11618027016247898,
|
| 2352 |
+
"grad_norm": 5.22974157333374,
|
| 2353 |
+
"learning_rate": 0.0002995866454052195,
|
| 2354 |
+
"loss": 6.560474395751953,
|
| 2355 |
+
"step": 3350
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.11652707693908339,
|
| 2359 |
+
"grad_norm": 4.654702663421631,
|
| 2360 |
+
"learning_rate": 0.000299583106551855,
|
| 2361 |
+
"loss": 6.996910095214844,
|
| 2362 |
+
"step": 3360
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.1168738837156878,
|
| 2366 |
+
"grad_norm": 7.279356479644775,
|
| 2367 |
+
"learning_rate": 0.00029957955263547575,
|
| 2368 |
+
"loss": 7.00561294555664,
|
| 2369 |
+
"step": 3370
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.11722069049229222,
|
| 2373 |
+
"grad_norm": 4.758134365081787,
|
| 2374 |
+
"learning_rate": 0.00029957598365643956,
|
| 2375 |
+
"loss": 7.142219543457031,
|
| 2376 |
+
"step": 3380
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.11756749726889663,
|
| 2380 |
+
"grad_norm": 5.935283660888672,
|
| 2381 |
+
"learning_rate": 0.000299572399615106,
|
| 2382 |
+
"loss": 6.959394073486328,
|
| 2383 |
+
"step": 3390
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.11791430404550104,
|
| 2387 |
+
"grad_norm": 5.490697383880615,
|
| 2388 |
+
"learning_rate": 0.00029956880051183586,
|
| 2389 |
+
"loss": 6.956050872802734,
|
| 2390 |
+
"step": 3400
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.11826111082210547,
|
| 2394 |
+
"grad_norm": 5.379637718200684,
|
| 2395 |
+
"learning_rate": 0.0002995651863469916,
|
| 2396 |
+
"loss": 7.505924987792969,
|
| 2397 |
+
"step": 3410
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.11860791759870988,
|
| 2401 |
+
"grad_norm": 4.9199934005737305,
|
| 2402 |
+
"learning_rate": 0.00029956155712093717,
|
| 2403 |
+
"loss": 7.151496887207031,
|
| 2404 |
+
"step": 3420
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.11895472437531429,
|
| 2408 |
+
"grad_norm": 5.5413103103637695,
|
| 2409 |
+
"learning_rate": 0.000299557912834038,
|
| 2410 |
+
"loss": 6.638466644287109,
|
| 2411 |
+
"step": 3430
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.1193015311519187,
|
| 2415 |
+
"grad_norm": 5.7927069664001465,
|
| 2416 |
+
"learning_rate": 0.00029955425348666113,
|
| 2417 |
+
"loss": 7.035697174072266,
|
| 2418 |
+
"step": 3440
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.11964833792852313,
|
| 2422 |
+
"grad_norm": 6.627818584442139,
|
| 2423 |
+
"learning_rate": 0.000299550579079175,
|
| 2424 |
+
"loss": 6.852809143066406,
|
| 2425 |
+
"step": 3450
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.11999514470512754,
|
| 2429 |
+
"grad_norm": 6.04095983505249,
|
| 2430 |
+
"learning_rate": 0.0002995468896119497,
|
| 2431 |
+
"loss": 6.934154510498047,
|
| 2432 |
+
"step": 3460
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.12034195148173195,
|
| 2436 |
+
"grad_norm": 6.020833969116211,
|
| 2437 |
+
"learning_rate": 0.0002995431850853567,
|
| 2438 |
+
"loss": 7.179639434814453,
|
| 2439 |
+
"step": 3470
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.12068875825833637,
|
| 2443 |
+
"grad_norm": 5.752357482910156,
|
| 2444 |
+
"learning_rate": 0.00029953946549976903,
|
| 2445 |
+
"loss": 6.6078651428222654,
|
| 2446 |
+
"step": 3480
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.12103556503494078,
|
| 2450 |
+
"grad_norm": 6.476692199707031,
|
| 2451 |
+
"learning_rate": 0.00029953573085556126,
|
| 2452 |
+
"loss": 7.1372932434082035,
|
| 2453 |
+
"step": 3490
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.1213823718115452,
|
| 2457 |
+
"grad_norm": 7.356377124786377,
|
| 2458 |
+
"learning_rate": 0.0002995319811531095,
|
| 2459 |
+
"loss": 7.113587188720703,
|
| 2460 |
+
"step": 3500
|
| 2461 |
+
}
|
| 2462 |
+
],
|
| 2463 |
+
"logging_steps": 10,
|
| 2464 |
+
"max_steps": 100000,
|
| 2465 |
+
"num_input_tokens_seen": 0,
|
| 2466 |
+
"num_train_epochs": 4,
|
| 2467 |
+
"save_steps": 100,
|
| 2468 |
+
"stateful_callbacks": {
|
| 2469 |
+
"TrainerControl": {
|
| 2470 |
+
"args": {
|
| 2471 |
+
"should_epoch_stop": false,
|
| 2472 |
+
"should_evaluate": false,
|
| 2473 |
+
"should_log": false,
|
| 2474 |
+
"should_save": true,
|
| 2475 |
+
"should_training_stop": false
|
| 2476 |
+
},
|
| 2477 |
+
"attributes": {}
|
| 2478 |
+
}
|
| 2479 |
+
},
|
| 2480 |
+
"total_flos": 2.1769697230848e+16,
|
| 2481 |
+
"train_batch_size": 4,
|
| 2482 |
+
"trial_name": null,
|
| 2483 |
+
"trial_params": null
|
| 2484 |
+
}
|
checkpoint-3500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cbd241599217f087cf90c5afde8d5857bb2c3234f97a197910829d91c3d8c1f7
|
| 3 |
+
size 5137
|