Upload folder using huggingface_hub
Browse files- checkpoint-1600/config.json +16 -0
- checkpoint-1600/generation_config.json +6 -0
- checkpoint-1600/model.safetensors +3 -0
- checkpoint-1600/optimizer.pt +3 -0
- checkpoint-1600/rng_state.pth +3 -0
- checkpoint-1600/scaler.pt +3 -0
- checkpoint-1600/scheduler.pt +3 -0
- checkpoint-1600/trainer_state.json +1154 -0
- checkpoint-1600/training_args.bin +3 -0
checkpoint-1600/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-1600/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-1600/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f2ea749066a4fe37d258b247657e17a7e5f1e736de40a04d32a63a41b4585fcd
|
| 3 |
+
size 543424784
|
checkpoint-1600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:692f88ddd36d0e750490ed09157c5954e1855e5b5579b333c6c33dffbdc91113
|
| 3 |
+
size 273342083
|
checkpoint-1600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea14519104a84f85cdfbb9352c81607bba94ec1ef2d199848562f341c5ab4236
|
| 3 |
+
size 14645
|
checkpoint-1600/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:90eadfc7339a8e0d1c29a03a4cf1b7a120e8d8fcf42e7b3f97e7b2c235cf286f
|
| 3 |
+
size 1383
|
checkpoint-1600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:766865e808b97ee264bc159c65aad661548471a27d0f786a41a84cea95e598a3
|
| 3 |
+
size 1465
|
checkpoint-1600/trainer_state.json
ADDED
|
@@ -0,0 +1,1154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.11097816851341276,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1600,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0006936135532088297,
|
| 14 |
+
"grad_norm": 80.97769927978516,
|
| 15 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 16 |
+
"loss": 39.45225219726562,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0013872271064176594,
|
| 21 |
+
"grad_norm": 69.19830322265625,
|
| 22 |
+
"learning_rate": 6.25e-06,
|
| 23 |
+
"loss": 38.231222534179686,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0020808406596264893,
|
| 28 |
+
"grad_norm": 55.03429412841797,
|
| 29 |
+
"learning_rate": 1.0416666666666666e-05,
|
| 30 |
+
"loss": 37.07829284667969,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.002774454212835319,
|
| 35 |
+
"grad_norm": 36.03839874267578,
|
| 36 |
+
"learning_rate": 1.4583333333333333e-05,
|
| 37 |
+
"loss": 35.945098876953125,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0034680677660441487,
|
| 42 |
+
"grad_norm": 48.63792419433594,
|
| 43 |
+
"learning_rate": 1.875e-05,
|
| 44 |
+
"loss": 34.40289306640625,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.004161681319252979,
|
| 49 |
+
"grad_norm": 41.98967361450195,
|
| 50 |
+
"learning_rate": 2.2916666666666667e-05,
|
| 51 |
+
"loss": 33.45003662109375,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.004855294872461808,
|
| 56 |
+
"grad_norm": 81.57576751708984,
|
| 57 |
+
"learning_rate": 2.7083333333333335e-05,
|
| 58 |
+
"loss": 32.02000427246094,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.005548908425670638,
|
| 63 |
+
"grad_norm": 60.575645446777344,
|
| 64 |
+
"learning_rate": 3.125e-05,
|
| 65 |
+
"loss": 30.74504699707031,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.006242521978879467,
|
| 70 |
+
"grad_norm": 61.149410247802734,
|
| 71 |
+
"learning_rate": 3.541666666666667e-05,
|
| 72 |
+
"loss": 29.49912109375,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.006936135532088297,
|
| 77 |
+
"grad_norm": 69.96782684326172,
|
| 78 |
+
"learning_rate": 3.958333333333333e-05,
|
| 79 |
+
"loss": 28.152670288085936,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.007629749085297127,
|
| 84 |
+
"grad_norm": 33.35784149169922,
|
| 85 |
+
"learning_rate": 4.375e-05,
|
| 86 |
+
"loss": 27.86654052734375,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.008323362638505957,
|
| 91 |
+
"grad_norm": 42.682308197021484,
|
| 92 |
+
"learning_rate": 4.791666666666667e-05,
|
| 93 |
+
"loss": 27.098068237304688,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.009016976191714786,
|
| 98 |
+
"grad_norm": 45.40013122558594,
|
| 99 |
+
"learning_rate": 5.208333333333334e-05,
|
| 100 |
+
"loss": 25.768472290039064,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.009710589744923616,
|
| 105 |
+
"grad_norm": 36.54875564575195,
|
| 106 |
+
"learning_rate": 5.6250000000000005e-05,
|
| 107 |
+
"loss": 25.419296264648438,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.010404203298132445,
|
| 112 |
+
"grad_norm": 54.964447021484375,
|
| 113 |
+
"learning_rate": 6.041666666666667e-05,
|
| 114 |
+
"loss": 24.5228759765625,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.011097816851341275,
|
| 119 |
+
"grad_norm": 62.00859451293945,
|
| 120 |
+
"learning_rate": 6.458333333333334e-05,
|
| 121 |
+
"loss": 23.92150115966797,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.011791430404550106,
|
| 126 |
+
"grad_norm": 50.303070068359375,
|
| 127 |
+
"learning_rate": 6.875e-05,
|
| 128 |
+
"loss": 23.61223449707031,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.012485043957758934,
|
| 133 |
+
"grad_norm": 40.088356018066406,
|
| 134 |
+
"learning_rate": 7.291666666666667e-05,
|
| 135 |
+
"loss": 23.29130401611328,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.013178657510967764,
|
| 140 |
+
"grad_norm": 58.86671829223633,
|
| 141 |
+
"learning_rate": 7.708333333333334e-05,
|
| 142 |
+
"loss": 23.32776794433594,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.013872271064176595,
|
| 147 |
+
"grad_norm": 38.86407470703125,
|
| 148 |
+
"learning_rate": 8.125000000000001e-05,
|
| 149 |
+
"loss": 23.280059814453125,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.014565884617385423,
|
| 154 |
+
"grad_norm": 37.631717681884766,
|
| 155 |
+
"learning_rate": 8.541666666666666e-05,
|
| 156 |
+
"loss": 22.259375,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.015259498170594254,
|
| 161 |
+
"grad_norm": 37.38259506225586,
|
| 162 |
+
"learning_rate": 8.958333333333333e-05,
|
| 163 |
+
"loss": 21.507147216796874,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.015953111723803082,
|
| 168 |
+
"grad_norm": 42.106143951416016,
|
| 169 |
+
"learning_rate": 9.375e-05,
|
| 170 |
+
"loss": 21.231875610351562,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.016646725277011915,
|
| 175 |
+
"grad_norm": 50.364681243896484,
|
| 176 |
+
"learning_rate": 9.791666666666667e-05,
|
| 177 |
+
"loss": 21.234603881835938,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.017340338830220743,
|
| 182 |
+
"grad_norm": 40.50483322143555,
|
| 183 |
+
"learning_rate": 0.00010208333333333334,
|
| 184 |
+
"loss": 20.62218017578125,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.018033952383429572,
|
| 189 |
+
"grad_norm": 60.68687057495117,
|
| 190 |
+
"learning_rate": 0.00010625,
|
| 191 |
+
"loss": 19.96156463623047,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.0187275659366384,
|
| 196 |
+
"grad_norm": 31.746686935424805,
|
| 197 |
+
"learning_rate": 0.00011041666666666666,
|
| 198 |
+
"loss": 19.283287048339844,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.019421179489847232,
|
| 203 |
+
"grad_norm": 40.83134841918945,
|
| 204 |
+
"learning_rate": 0.00011458333333333333,
|
| 205 |
+
"loss": 18.95526580810547,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.02011479304305606,
|
| 210 |
+
"grad_norm": 27.94510841369629,
|
| 211 |
+
"learning_rate": 0.00011875,
|
| 212 |
+
"loss": 19.398623657226562,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.02080840659626489,
|
| 217 |
+
"grad_norm": 43.998260498046875,
|
| 218 |
+
"learning_rate": 0.00012291666666666665,
|
| 219 |
+
"loss": 18.262245178222656,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.021502020149473722,
|
| 224 |
+
"grad_norm": 22.61271095275879,
|
| 225 |
+
"learning_rate": 0.00012708333333333332,
|
| 226 |
+
"loss": 18.646043395996095,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.02219563370268255,
|
| 231 |
+
"grad_norm": 27.580198287963867,
|
| 232 |
+
"learning_rate": 0.00013125000000000002,
|
| 233 |
+
"loss": 17.519342041015626,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.02288924725589138,
|
| 238 |
+
"grad_norm": 36.54242706298828,
|
| 239 |
+
"learning_rate": 0.00013541666666666666,
|
| 240 |
+
"loss": 16.588734436035157,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.02358286080910021,
|
| 245 |
+
"grad_norm": 24.73045539855957,
|
| 246 |
+
"learning_rate": 0.00013958333333333333,
|
| 247 |
+
"loss": 16.85819091796875,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.02427647436230904,
|
| 252 |
+
"grad_norm": 27.577693939208984,
|
| 253 |
+
"learning_rate": 0.00014375,
|
| 254 |
+
"loss": 16.190365600585938,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.02497008791551787,
|
| 259 |
+
"grad_norm": 20.408132553100586,
|
| 260 |
+
"learning_rate": 0.00014791666666666667,
|
| 261 |
+
"loss": 15.682986450195312,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.0256637014687267,
|
| 266 |
+
"grad_norm": 28.981483459472656,
|
| 267 |
+
"learning_rate": 0.00015208333333333333,
|
| 268 |
+
"loss": 15.275959777832032,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.02635731502193553,
|
| 273 |
+
"grad_norm": 19.978363037109375,
|
| 274 |
+
"learning_rate": 0.00015625,
|
| 275 |
+
"loss": 15.033628845214844,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.027050928575144358,
|
| 280 |
+
"grad_norm": 15.66555404663086,
|
| 281 |
+
"learning_rate": 0.00016041666666666667,
|
| 282 |
+
"loss": 14.446699523925782,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.02774454212835319,
|
| 287 |
+
"grad_norm": 17.955888748168945,
|
| 288 |
+
"learning_rate": 0.00016458333333333334,
|
| 289 |
+
"loss": 14.196858215332032,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.02843815568156202,
|
| 294 |
+
"grad_norm": 16.955345153808594,
|
| 295 |
+
"learning_rate": 0.00016875,
|
| 296 |
+
"loss": 13.838204956054687,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.029131769234770847,
|
| 301 |
+
"grad_norm": 14.126781463623047,
|
| 302 |
+
"learning_rate": 0.00017291666666666668,
|
| 303 |
+
"loss": 13.345814514160157,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.029825382787979676,
|
| 308 |
+
"grad_norm": 16.29298973083496,
|
| 309 |
+
"learning_rate": 0.00017708333333333335,
|
| 310 |
+
"loss": 13.06510009765625,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.030518996341188508,
|
| 315 |
+
"grad_norm": 14.477455139160156,
|
| 316 |
+
"learning_rate": 0.00018125,
|
| 317 |
+
"loss": 12.520622253417969,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.031212609894397336,
|
| 322 |
+
"grad_norm": 12.074920654296875,
|
| 323 |
+
"learning_rate": 0.00018541666666666668,
|
| 324 |
+
"loss": 12.411215209960938,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.031906223447606165,
|
| 329 |
+
"grad_norm": 15.564898490905762,
|
| 330 |
+
"learning_rate": 0.00018958333333333332,
|
| 331 |
+
"loss": 12.00233154296875,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.032599837000814993,
|
| 336 |
+
"grad_norm": 15.621328353881836,
|
| 337 |
+
"learning_rate": 0.00019375000000000002,
|
| 338 |
+
"loss": 12.109133911132812,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.03329345055402383,
|
| 343 |
+
"grad_norm": 10.823607444763184,
|
| 344 |
+
"learning_rate": 0.00019791666666666666,
|
| 345 |
+
"loss": 11.684900665283203,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.03398706410723266,
|
| 350 |
+
"grad_norm": 14.166363716125488,
|
| 351 |
+
"learning_rate": 0.00020208333333333333,
|
| 352 |
+
"loss": 11.60366439819336,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.034680677660441486,
|
| 357 |
+
"grad_norm": 39.65184783935547,
|
| 358 |
+
"learning_rate": 0.00020625,
|
| 359 |
+
"loss": 11.060487365722656,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.035374291213650315,
|
| 364 |
+
"grad_norm": 12.022449493408203,
|
| 365 |
+
"learning_rate": 0.00021041666666666667,
|
| 366 |
+
"loss": 11.373404693603515,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.036067904766859143,
|
| 371 |
+
"grad_norm": 9.625805854797363,
|
| 372 |
+
"learning_rate": 0.00021458333333333334,
|
| 373 |
+
"loss": 11.405331420898438,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.03676151832006797,
|
| 378 |
+
"grad_norm": 12.523700714111328,
|
| 379 |
+
"learning_rate": 0.00021875,
|
| 380 |
+
"loss": 10.810578155517579,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.0374551318732768,
|
| 385 |
+
"grad_norm": 11.111766815185547,
|
| 386 |
+
"learning_rate": 0.00022291666666666667,
|
| 387 |
+
"loss": 10.427848815917969,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.038148745426485636,
|
| 392 |
+
"grad_norm": 9.552438735961914,
|
| 393 |
+
"learning_rate": 0.00022708333333333334,
|
| 394 |
+
"loss": 10.221891784667969,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.038842358979694465,
|
| 399 |
+
"grad_norm": 10.854446411132812,
|
| 400 |
+
"learning_rate": 0.00023125,
|
| 401 |
+
"loss": 10.453067779541016,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.039535972532903293,
|
| 406 |
+
"grad_norm": 8.054693222045898,
|
| 407 |
+
"learning_rate": 0.00023541666666666668,
|
| 408 |
+
"loss": 9.835127258300782,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.04022958608611212,
|
| 413 |
+
"grad_norm": 9.217058181762695,
|
| 414 |
+
"learning_rate": 0.00023958333333333335,
|
| 415 |
+
"loss": 9.983591461181641,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.04092319963932095,
|
| 420 |
+
"grad_norm": 7.442865371704102,
|
| 421 |
+
"learning_rate": 0.00024375,
|
| 422 |
+
"loss": 9.785844421386718,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.04161681319252978,
|
| 427 |
+
"grad_norm": 7.499183177947998,
|
| 428 |
+
"learning_rate": 0.0002479166666666667,
|
| 429 |
+
"loss": 9.547152709960937,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.042310426745738615,
|
| 434 |
+
"grad_norm": 6.999185085296631,
|
| 435 |
+
"learning_rate": 0.0002520833333333333,
|
| 436 |
+
"loss": 9.602791595458985,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.043004040298947444,
|
| 441 |
+
"grad_norm": 8.540186882019043,
|
| 442 |
+
"learning_rate": 0.00025624999999999997,
|
| 443 |
+
"loss": 9.688284301757813,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.04369765385215627,
|
| 448 |
+
"grad_norm": 7.028185844421387,
|
| 449 |
+
"learning_rate": 0.0002604166666666667,
|
| 450 |
+
"loss": 9.46673355102539,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.0443912674053651,
|
| 455 |
+
"grad_norm": 10.2844820022583,
|
| 456 |
+
"learning_rate": 0.00026458333333333336,
|
| 457 |
+
"loss": 9.028829956054688,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.04508488095857393,
|
| 462 |
+
"grad_norm": 6.920653820037842,
|
| 463 |
+
"learning_rate": 0.00026875,
|
| 464 |
+
"loss": 9.584944152832032,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.04577849451178276,
|
| 469 |
+
"grad_norm": 7.473561763763428,
|
| 470 |
+
"learning_rate": 0.00027291666666666664,
|
| 471 |
+
"loss": 9.417964935302734,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.04647210806499159,
|
| 476 |
+
"grad_norm": 6.416537761688232,
|
| 477 |
+
"learning_rate": 0.00027708333333333334,
|
| 478 |
+
"loss": 9.235848236083985,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.04716572161820042,
|
| 483 |
+
"grad_norm": 7.264041900634766,
|
| 484 |
+
"learning_rate": 0.00028125000000000003,
|
| 485 |
+
"loss": 9.052194976806641,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.04785933517140925,
|
| 490 |
+
"grad_norm": 5.6128692626953125,
|
| 491 |
+
"learning_rate": 0.0002854166666666667,
|
| 492 |
+
"loss": 8.73398208618164,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.04855294872461808,
|
| 497 |
+
"grad_norm": 6.392711639404297,
|
| 498 |
+
"learning_rate": 0.00028958333333333337,
|
| 499 |
+
"loss": 8.775264739990234,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.04924656227782691,
|
| 504 |
+
"grad_norm": 7.4516496658325195,
|
| 505 |
+
"learning_rate": 0.00029375,
|
| 506 |
+
"loss": 8.859925079345704,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.04994017583103574,
|
| 511 |
+
"grad_norm": 6.566150188446045,
|
| 512 |
+
"learning_rate": 0.00029791666666666665,
|
| 513 |
+
"loss": 9.032890319824219,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.050633789384244565,
|
| 518 |
+
"grad_norm": 7.080583572387695,
|
| 519 |
+
"learning_rate": 0.0003020833333333333,
|
| 520 |
+
"loss": 8.99814682006836,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.0513274029374534,
|
| 525 |
+
"grad_norm": 5.690314292907715,
|
| 526 |
+
"learning_rate": 0.00030625000000000004,
|
| 527 |
+
"loss": 8.487692260742188,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.05202101649066223,
|
| 532 |
+
"grad_norm": 6.124709606170654,
|
| 533 |
+
"learning_rate": 0.0003104166666666667,
|
| 534 |
+
"loss": 9.122718811035156,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.05271463004387106,
|
| 539 |
+
"grad_norm": 5.513874530792236,
|
| 540 |
+
"learning_rate": 0.00031458333333333333,
|
| 541 |
+
"loss": 8.884898376464843,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.05340824359707989,
|
| 546 |
+
"grad_norm": 6.077105522155762,
|
| 547 |
+
"learning_rate": 0.00031874999999999997,
|
| 548 |
+
"loss": 8.678743743896485,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.054101857150288715,
|
| 553 |
+
"grad_norm": 5.43304967880249,
|
| 554 |
+
"learning_rate": 0.0003229166666666667,
|
| 555 |
+
"loss": 8.514462280273438,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.054795470703497544,
|
| 560 |
+
"grad_norm": 6.31812858581543,
|
| 561 |
+
"learning_rate": 0.00032708333333333336,
|
| 562 |
+
"loss": 8.414434814453125,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.05548908425670638,
|
| 567 |
+
"grad_norm": 6.50822114944458,
|
| 568 |
+
"learning_rate": 0.00033125,
|
| 569 |
+
"loss": 8.508963775634765,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.05618269780991521,
|
| 574 |
+
"grad_norm": 5.709455966949463,
|
| 575 |
+
"learning_rate": 0.00033541666666666664,
|
| 576 |
+
"loss": 8.19250030517578,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.05687631136312404,
|
| 581 |
+
"grad_norm": 6.013482093811035,
|
| 582 |
+
"learning_rate": 0.00033958333333333334,
|
| 583 |
+
"loss": 8.754444885253907,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.057569924916332865,
|
| 588 |
+
"grad_norm": 6.8106842041015625,
|
| 589 |
+
"learning_rate": 0.00034375,
|
| 590 |
+
"loss": 8.26578598022461,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.058263538469541694,
|
| 595 |
+
"grad_norm": 5.135249614715576,
|
| 596 |
+
"learning_rate": 0.0003479166666666667,
|
| 597 |
+
"loss": 8.549038696289063,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.05895715202275052,
|
| 602 |
+
"grad_norm": 5.1326823234558105,
|
| 603 |
+
"learning_rate": 0.00035208333333333337,
|
| 604 |
+
"loss": 8.302710723876952,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.05965076557595935,
|
| 609 |
+
"grad_norm": 4.964430809020996,
|
| 610 |
+
"learning_rate": 0.00035625,
|
| 611 |
+
"loss": 8.092277526855469,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.06034437912916819,
|
| 616 |
+
"grad_norm": 6.6412882804870605,
|
| 617 |
+
"learning_rate": 0.00036041666666666665,
|
| 618 |
+
"loss": 7.938130187988281,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.061037992682377015,
|
| 623 |
+
"grad_norm": 5.193583011627197,
|
| 624 |
+
"learning_rate": 0.0003645833333333333,
|
| 625 |
+
"loss": 8.546875,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.061731606235585844,
|
| 630 |
+
"grad_norm": 4.927044868469238,
|
| 631 |
+
"learning_rate": 0.00036875000000000005,
|
| 632 |
+
"loss": 7.911424255371093,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.06242521978879467,
|
| 637 |
+
"grad_norm": 5.043755531311035,
|
| 638 |
+
"learning_rate": 0.0003729166666666667,
|
| 639 |
+
"loss": 8.359856414794923,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.06311883334200351,
|
| 644 |
+
"grad_norm": 5.2342352867126465,
|
| 645 |
+
"learning_rate": 0.00037708333333333333,
|
| 646 |
+
"loss": 7.912593841552734,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.06381244689521233,
|
| 651 |
+
"grad_norm": 4.811470985412598,
|
| 652 |
+
"learning_rate": 0.00038124999999999997,
|
| 653 |
+
"loss": 8.026778411865234,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.06450606044842117,
|
| 658 |
+
"grad_norm": 5.230201721191406,
|
| 659 |
+
"learning_rate": 0.0003854166666666667,
|
| 660 |
+
"loss": 7.771602630615234,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.06519967400162999,
|
| 665 |
+
"grad_norm": 5.551824569702148,
|
| 666 |
+
"learning_rate": 0.00038958333333333336,
|
| 667 |
+
"loss": 7.761222839355469,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.06589328755483882,
|
| 672 |
+
"grad_norm": 4.743410110473633,
|
| 673 |
+
"learning_rate": 0.00039375,
|
| 674 |
+
"loss": 7.748975372314453,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.06658690110804766,
|
| 679 |
+
"grad_norm": 5.341291427612305,
|
| 680 |
+
"learning_rate": 0.00039791666666666664,
|
| 681 |
+
"loss": 7.9248199462890625,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.06728051466125648,
|
| 686 |
+
"grad_norm": 4.554710865020752,
|
| 687 |
+
"learning_rate": 0.00040166666666666665,
|
| 688 |
+
"loss": 7.922956085205078,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.06797412821446532,
|
| 693 |
+
"grad_norm": 4.599163055419922,
|
| 694 |
+
"learning_rate": 0.00040583333333333335,
|
| 695 |
+
"loss": 7.850189208984375,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.06866774176767414,
|
| 700 |
+
"grad_norm": 4.907506942749023,
|
| 701 |
+
"learning_rate": 0.00041,
|
| 702 |
+
"loss": 7.55450668334961,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.06936135532088297,
|
| 707 |
+
"grad_norm": 4.697291374206543,
|
| 708 |
+
"learning_rate": 0.0004141666666666667,
|
| 709 |
+
"loss": 7.54024658203125,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.0700549688740918,
|
| 714 |
+
"grad_norm": 5.102283000946045,
|
| 715 |
+
"learning_rate": 0.00041833333333333333,
|
| 716 |
+
"loss": 7.947976684570312,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.07074858242730063,
|
| 721 |
+
"grad_norm": 5.308175563812256,
|
| 722 |
+
"learning_rate": 0.00042249999999999997,
|
| 723 |
+
"loss": 7.837068176269531,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.07144219598050947,
|
| 728 |
+
"grad_norm": 5.058789253234863,
|
| 729 |
+
"learning_rate": 0.0004266666666666667,
|
| 730 |
+
"loss": 7.611346435546875,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.07213580953371829,
|
| 735 |
+
"grad_norm": 5.363205432891846,
|
| 736 |
+
"learning_rate": 0.00043083333333333336,
|
| 737 |
+
"loss": 7.675627899169922,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.07282942308692712,
|
| 742 |
+
"grad_norm": 5.060066223144531,
|
| 743 |
+
"learning_rate": 0.000435,
|
| 744 |
+
"loss": 7.799678802490234,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.07352303664013594,
|
| 749 |
+
"grad_norm": 5.268683433532715,
|
| 750 |
+
"learning_rate": 0.00043916666666666664,
|
| 751 |
+
"loss": 7.417415618896484,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.07421665019334478,
|
| 756 |
+
"grad_norm": 29.964582443237305,
|
| 757 |
+
"learning_rate": 0.00044333333333333334,
|
| 758 |
+
"loss": 7.611949920654297,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.0749102637465536,
|
| 763 |
+
"grad_norm": 6.951490879058838,
|
| 764 |
+
"learning_rate": 0.00044750000000000004,
|
| 765 |
+
"loss": 7.826107025146484,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.07560387729976244,
|
| 770 |
+
"grad_norm": Infinity,
|
| 771 |
+
"learning_rate": 0.0004516666666666667,
|
| 772 |
+
"loss": 8.342691040039062,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.07629749085297127,
|
| 777 |
+
"grad_norm": 19.79059410095215,
|
| 778 |
+
"learning_rate": 0.0004554166666666667,
|
| 779 |
+
"loss": 8.860975646972657,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.0769911044061801,
|
| 784 |
+
"grad_norm": 5.309981346130371,
|
| 785 |
+
"learning_rate": 0.00045958333333333333,
|
| 786 |
+
"loss": 8.495597839355469,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.07768471795938893,
|
| 791 |
+
"grad_norm": 62.24746322631836,
|
| 792 |
+
"learning_rate": 0.00046375,
|
| 793 |
+
"loss": 7.669882965087891,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.07837833151259775,
|
| 798 |
+
"grad_norm": Infinity,
|
| 799 |
+
"learning_rate": 0.00046791666666666667,
|
| 800 |
+
"loss": 9.232691192626953,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.07907194506580659,
|
| 805 |
+
"grad_norm": 21.891761779785156,
|
| 806 |
+
"learning_rate": 0.00047125,
|
| 807 |
+
"loss": 9.408066558837891,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.07976555861901541,
|
| 812 |
+
"grad_norm": 47.06490707397461,
|
| 813 |
+
"learning_rate": 0.0004754166666666667,
|
| 814 |
+
"loss": 9.484445190429687,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.08045917217222424,
|
| 819 |
+
"grad_norm": 34.64010238647461,
|
| 820 |
+
"learning_rate": 0.0004795833333333333,
|
| 821 |
+
"loss": 9.570259857177735,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.08115278572543308,
|
| 826 |
+
"grad_norm": 12.11959457397461,
|
| 827 |
+
"learning_rate": 0.00048375,
|
| 828 |
+
"loss": 9.447135162353515,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.0818463992786419,
|
| 833 |
+
"grad_norm": 16.15092658996582,
|
| 834 |
+
"learning_rate": 0.0004875,
|
| 835 |
+
"loss": 9.065760040283203,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.08254001283185074,
|
| 840 |
+
"grad_norm": 157.4208526611328,
|
| 841 |
+
"learning_rate": 0.0004916666666666666,
|
| 842 |
+
"loss": 8.82796630859375,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.08323362638505956,
|
| 847 |
+
"grad_norm": 27.526687622070312,
|
| 848 |
+
"learning_rate": 0.0004958333333333334,
|
| 849 |
+
"loss": 8.742295837402343,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.0839272399382684,
|
| 854 |
+
"grad_norm": 145.7564697265625,
|
| 855 |
+
"learning_rate": 0.0005,
|
| 856 |
+
"loss": 8.395146942138672,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.08462085349147723,
|
| 861 |
+
"grad_norm": 6.748693466186523,
|
| 862 |
+
"learning_rate": 0.0004999999873614903,
|
| 863 |
+
"loss": 8.451645660400391,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.08531446704468605,
|
| 868 |
+
"grad_norm": 22.430376052856445,
|
| 869 |
+
"learning_rate": 0.0004999999494459626,
|
| 870 |
+
"loss": 8.06414794921875,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.08600808059789489,
|
| 875 |
+
"grad_norm": 11.808032035827637,
|
| 876 |
+
"learning_rate": 0.0004999998862534205,
|
| 877 |
+
"loss": 8.445089721679688,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.08670169415110371,
|
| 882 |
+
"grad_norm": 40.87104034423828,
|
| 883 |
+
"learning_rate": 0.0004999997977838706,
|
| 884 |
+
"loss": 8.36043930053711,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.08739530770431254,
|
| 889 |
+
"grad_norm": 6.653709411621094,
|
| 890 |
+
"learning_rate": 0.0004999996840373218,
|
| 891 |
+
"loss": 8.253304290771485,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.08808892125752137,
|
| 896 |
+
"grad_norm": 143.88558959960938,
|
| 897 |
+
"learning_rate": 0.0004999995450137855,
|
| 898 |
+
"loss": 8.144001770019532,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.0887825348107302,
|
| 903 |
+
"grad_norm": 21.387109756469727,
|
| 904 |
+
"learning_rate": 0.0004999993807132759,
|
| 905 |
+
"loss": 8.308104705810546,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.08947614836393904,
|
| 910 |
+
"grad_norm": 24.346641540527344,
|
| 911 |
+
"learning_rate": 0.0004999991911358096,
|
| 912 |
+
"loss": 8.206471252441407,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.09016976191714786,
|
| 917 |
+
"grad_norm": 5.9509124755859375,
|
| 918 |
+
"learning_rate": 0.0004999989762814056,
|
| 919 |
+
"loss": 7.916456604003907,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.0908633754703567,
|
| 924 |
+
"grad_norm": 4.6718292236328125,
|
| 925 |
+
"learning_rate": 0.0004999987361500857,
|
| 926 |
+
"loss": 8.069509887695313,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.09155698902356552,
|
| 931 |
+
"grad_norm": 7.908266067504883,
|
| 932 |
+
"learning_rate": 0.0004999984707418744,
|
| 933 |
+
"loss": 7.811135101318359,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.09225060257677435,
|
| 938 |
+
"grad_norm": 5.006199836730957,
|
| 939 |
+
"learning_rate": 0.0004999981800567982,
|
| 940 |
+
"loss": 7.62738037109375,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.09294421612998317,
|
| 945 |
+
"grad_norm": 5.048142433166504,
|
| 946 |
+
"learning_rate": 0.0004999979293094143,
|
| 947 |
+
"loss": 7.573948669433594,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.09363782968319201,
|
| 952 |
+
"grad_norm": 315.8863220214844,
|
| 953 |
+
"learning_rate": 0.0004999975931260575,
|
| 954 |
+
"loss": 7.884748077392578,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.09433144323640084,
|
| 959 |
+
"grad_norm": 6.798584938049316,
|
| 960 |
+
"learning_rate": 0.0004999972689493917,
|
| 961 |
+
"loss": 8.001516723632813,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.09502505678960967,
|
| 966 |
+
"grad_norm": 1015.7785034179688,
|
| 967 |
+
"learning_rate": 0.0004999968847401915,
|
| 968 |
+
"loss": 7.8455970764160154,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.0957186703428185,
|
| 973 |
+
"grad_norm": 10.685717582702637,
|
| 974 |
+
"learning_rate": 0.000499996475254287,
|
| 975 |
+
"loss": 8.098987579345703,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.09641228389602732,
|
| 980 |
+
"grad_norm": 11.820308685302734,
|
| 981 |
+
"learning_rate": 0.0004999960404917194,
|
| 982 |
+
"loss": 8.00228271484375,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.09710589744923616,
|
| 987 |
+
"grad_norm": 14.156989097595215,
|
| 988 |
+
"learning_rate": 0.0004999955804525329,
|
| 989 |
+
"loss": 7.7231597900390625,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.097799511002445,
|
| 994 |
+
"grad_norm": 14.044855117797852,
|
| 995 |
+
"learning_rate": 0.0004999950951367737,
|
| 996 |
+
"loss": 7.547687530517578,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.09849312455565382,
|
| 1001 |
+
"grad_norm": 21.774089813232422,
|
| 1002 |
+
"learning_rate": 0.0004999945845444912,
|
| 1003 |
+
"loss": 7.418048095703125,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.09918673810886265,
|
| 1008 |
+
"grad_norm": 24.606578826904297,
|
| 1009 |
+
"learning_rate": 0.0004999940486757369,
|
| 1010 |
+
"loss": 7.59954605102539,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.09988035166207147,
|
| 1015 |
+
"grad_norm": 17.820003509521484,
|
| 1016 |
+
"learning_rate": 0.0004999934875305649,
|
| 1017 |
+
"loss": 7.650421142578125,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.10057396521528031,
|
| 1022 |
+
"grad_norm": 23.51424789428711,
|
| 1023 |
+
"learning_rate": 0.000499992901109032,
|
| 1024 |
+
"loss": 8.069045257568359,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.10126757876848913,
|
| 1029 |
+
"grad_norm": 35.154178619384766,
|
| 1030 |
+
"learning_rate": 0.0004999922894111975,
|
| 1031 |
+
"loss": 7.217467498779297,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.10196119232169797,
|
| 1036 |
+
"grad_norm": 14.090352058410645,
|
| 1037 |
+
"learning_rate": 0.0004999916524371232,
|
| 1038 |
+
"loss": 7.3218132019042965,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.1026548058749068,
|
| 1043 |
+
"grad_norm": 22.718524932861328,
|
| 1044 |
+
"learning_rate": 0.0004999909901868735,
|
| 1045 |
+
"loss": 7.336325836181641,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.10334841942811562,
|
| 1050 |
+
"grad_norm": 24.450363159179688,
|
| 1051 |
+
"learning_rate": 0.0004999903026605154,
|
| 1052 |
+
"loss": 7.122545623779297,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.10404203298132446,
|
| 1057 |
+
"grad_norm": 19.112335205078125,
|
| 1058 |
+
"learning_rate": 0.0004999895898581185,
|
| 1059 |
+
"loss": 7.021141052246094,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.10473564653453328,
|
| 1064 |
+
"grad_norm": 37.39822769165039,
|
| 1065 |
+
"learning_rate": 0.0004999888517797547,
|
| 1066 |
+
"loss": 7.5175323486328125,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.10542926008774212,
|
| 1071 |
+
"grad_norm": 17.441190719604492,
|
| 1072 |
+
"learning_rate": 0.0004999880884254987,
|
| 1073 |
+
"loss": 7.584983825683594,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.10612287364095094,
|
| 1078 |
+
"grad_norm": 32.337440490722656,
|
| 1079 |
+
"learning_rate": 0.0004999872997954278,
|
| 1080 |
+
"loss": 7.30379638671875,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.10681648719415977,
|
| 1085 |
+
"grad_norm": 15.138467788696289,
|
| 1086 |
+
"learning_rate": 0.0004999864858896215,
|
| 1087 |
+
"loss": 7.259065246582031,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.10751010074736861,
|
| 1092 |
+
"grad_norm": 27.0513916015625,
|
| 1093 |
+
"learning_rate": 0.0004999856467081623,
|
| 1094 |
+
"loss": 7.271750640869141,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.10820371430057743,
|
| 1099 |
+
"grad_norm": 26.314247131347656,
|
| 1100 |
+
"learning_rate": 0.0004999847822511349,
|
| 1101 |
+
"loss": 7.2289573669433596,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.10889732785378627,
|
| 1106 |
+
"grad_norm": 22.641141891479492,
|
| 1107 |
+
"learning_rate": 0.0004999838925186269,
|
| 1108 |
+
"loss": 7.1235198974609375,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.10959094140699509,
|
| 1113 |
+
"grad_norm": 17.976776123046875,
|
| 1114 |
+
"learning_rate": 0.000499982977510728,
|
| 1115 |
+
"loss": 7.235462951660156,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.11028455496020392,
|
| 1120 |
+
"grad_norm": 20.597803115844727,
|
| 1121 |
+
"learning_rate": 0.0004999820372275308,
|
| 1122 |
+
"loss": 7.060716247558593,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.11097816851341276,
|
| 1127 |
+
"grad_norm": 24.85616683959961,
|
| 1128 |
+
"learning_rate": 0.0004999810716691305,
|
| 1129 |
+
"loss": 7.108528137207031,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
}
|
| 1132 |
+
],
|
| 1133 |
+
"logging_steps": 10,
|
| 1134 |
+
"max_steps": 100000,
|
| 1135 |
+
"num_input_tokens_seen": 0,
|
| 1136 |
+
"num_train_epochs": 7,
|
| 1137 |
+
"save_steps": 100,
|
| 1138 |
+
"stateful_callbacks": {
|
| 1139 |
+
"TrainerControl": {
|
| 1140 |
+
"args": {
|
| 1141 |
+
"should_epoch_stop": false,
|
| 1142 |
+
"should_evaluate": false,
|
| 1143 |
+
"should_log": false,
|
| 1144 |
+
"should_save": true,
|
| 1145 |
+
"should_training_stop": false
|
| 1146 |
+
},
|
| 1147 |
+
"attributes": {}
|
| 1148 |
+
}
|
| 1149 |
+
},
|
| 1150 |
+
"total_flos": 1.731588545421312e+16,
|
| 1151 |
+
"train_batch_size": 8,
|
| 1152 |
+
"trial_name": null,
|
| 1153 |
+
"trial_params": null
|
| 1154 |
+
}
|
checkpoint-1600/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49f3e14a77c081da7c71dab3b3b7c4ec1dede30e28c9fb753d87e3fc83f88ec1
|
| 3 |
+
size 5137
|