Upload folder using huggingface_hub
Browse files- checkpoint-2900/config.json +17 -0
- checkpoint-2900/generation_config.json +6 -0
- checkpoint-2900/hybrid_tokenizer_mapping.json +0 -0
- checkpoint-2900/model.safetensors +3 -0
- checkpoint-2900/optimizer.pt +3 -0
- checkpoint-2900/rng_state.pth +3 -0
- checkpoint-2900/scaler.pt +3 -0
- checkpoint-2900/scheduler.pt +3 -0
- checkpoint-2900/tokenizer_config.json +9 -0
- checkpoint-2900/trainer_state.json +2064 -0
- checkpoint-2900/training_args.bin +3 -0
checkpoint-2900/config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"FastHyperNet125mForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "float32",
|
| 6 |
+
"hidden_size": 768,
|
| 7 |
+
"hyperbolic_dim": 16,
|
| 8 |
+
"initializer_range": 0.02,
|
| 9 |
+
"intermediate_size": 2048,
|
| 10 |
+
"max_position_embeddings": 512,
|
| 11 |
+
"model_type": "fasthypernet_125m_tagless",
|
| 12 |
+
"num_attention_heads": 12,
|
| 13 |
+
"num_hidden_layers": 16,
|
| 14 |
+
"transformers_version": "5.13.1",
|
| 15 |
+
"use_cache": false,
|
| 16 |
+
"vocab_size": 8192
|
| 17 |
+
}
|
checkpoint-2900/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-2900/hybrid_tokenizer_mapping.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2900/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4e92922262ec6a759657f1e0a2405193253cc611d7162f806a0d1c90b2a57a8
|
| 3 |
+
size 501755480
|
checkpoint-2900/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c409e443d35d3eccf1f07a766ef04bb88d954497cfe317cf91030e301afbc14e
|
| 3 |
+
size 254410983
|
checkpoint-2900/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:377780e4d859ad8a8110ae47acb3d0aa57cabc718b2c42cb546cccd792e15ae8
|
| 3 |
+
size 14645
|
checkpoint-2900/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:adf752c0d05ffa4426054364cbb65c6a98a075f3a05882d6224c9b40928ad03f
|
| 3 |
+
size 1383
|
checkpoint-2900/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:043d4b18e73b2360c4f8c3406b8db925206f0f06b178c2d9021f4309f897c1ac
|
| 3 |
+
size 1465
|
checkpoint-2900/tokenizer_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"tokenizer_class": "HFCompatibleHybridTokenizer",
|
| 3 |
+
"vocab_size": 8192,
|
| 4 |
+
"model_max_length": 512,
|
| 5 |
+
"bos_token": "<bos>",
|
| 6 |
+
"eos_token": "<eos>",
|
| 7 |
+
"unk_token": "<unk>",
|
| 8 |
+
"pad_token": "<pad>"
|
| 9 |
+
}
|
checkpoint-2900/trainer_state.json
ADDED
|
@@ -0,0 +1,2064 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.005249688299757202,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2900,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0005249688299757202,
|
| 14 |
+
"grad_norm": 4051.18359375,
|
| 15 |
+
"learning_rate": 9e-07,
|
| 16 |
+
"loss": 7.93,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0010499376599514404,
|
| 21 |
+
"grad_norm": 111.26915740966797,
|
| 22 |
+
"learning_rate": 3.899999999999999e-06,
|
| 23 |
+
"loss": 7.7273,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0015749064899271605,
|
| 28 |
+
"grad_norm": 176.40713500976562,
|
| 29 |
+
"learning_rate": 6.899999999999999e-06,
|
| 30 |
+
"loss": 7.5271,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.002099875319902881,
|
| 35 |
+
"grad_norm": 668.3442993164062,
|
| 36 |
+
"learning_rate": 9.9e-06,
|
| 37 |
+
"loss": 7.6155,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.002624844149878601,
|
| 42 |
+
"grad_norm": 171.85369873046875,
|
| 43 |
+
"learning_rate": 1.2899999999999998e-05,
|
| 44 |
+
"loss": 7.4873,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.003149812979854321,
|
| 49 |
+
"grad_norm": 477.1031494140625,
|
| 50 |
+
"learning_rate": 1.5899999999999997e-05,
|
| 51 |
+
"loss": 7.4295,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.0036747818098300414,
|
| 56 |
+
"grad_norm": 256.7841491699219,
|
| 57 |
+
"learning_rate": 1.89e-05,
|
| 58 |
+
"loss": 7.3037,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.004199750639805762,
|
| 63 |
+
"grad_norm": 240.86167907714844,
|
| 64 |
+
"learning_rate": 2.1899999999999997e-05,
|
| 65 |
+
"loss": 7.3488,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0047247194697814816,
|
| 70 |
+
"grad_norm": 151.8752899169922,
|
| 71 |
+
"learning_rate": 2.49e-05,
|
| 72 |
+
"loss": 7.1664,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.005249688299757202,
|
| 77 |
+
"grad_norm": 92.41702270507812,
|
| 78 |
+
"learning_rate": 2.7899999999999997e-05,
|
| 79 |
+
"loss": 7.2001,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.005774657129732922,
|
| 84 |
+
"grad_norm": 125.6628189086914,
|
| 85 |
+
"learning_rate": 3.09e-05,
|
| 86 |
+
"loss": 7.1956,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.006299625959708642,
|
| 91 |
+
"grad_norm": 105.1700210571289,
|
| 92 |
+
"learning_rate": 3.39e-05,
|
| 93 |
+
"loss": 7.2136,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.006824594789684363,
|
| 98 |
+
"grad_norm": 95.49678039550781,
|
| 99 |
+
"learning_rate": 3.6899999999999996e-05,
|
| 100 |
+
"loss": 7.2046,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.007349563619660083,
|
| 105 |
+
"grad_norm": 70.33267974853516,
|
| 106 |
+
"learning_rate": 3.99e-05,
|
| 107 |
+
"loss": 7.0895,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.007874532449635803,
|
| 112 |
+
"grad_norm": 42.994972229003906,
|
| 113 |
+
"learning_rate": 4.289999999999999e-05,
|
| 114 |
+
"loss": 7.1157,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.008399501279611523,
|
| 119 |
+
"grad_norm": 87.23907470703125,
|
| 120 |
+
"learning_rate": 4.59e-05,
|
| 121 |
+
"loss": 7.1537,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.008924470109587244,
|
| 126 |
+
"grad_norm": 53.14045333862305,
|
| 127 |
+
"learning_rate": 4.8899999999999996e-05,
|
| 128 |
+
"loss": 7.0126,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.009449438939562963,
|
| 133 |
+
"grad_norm": 44.222225189208984,
|
| 134 |
+
"learning_rate": 5.1899999999999994e-05,
|
| 135 |
+
"loss": 7.1067,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.009974407769538684,
|
| 140 |
+
"grad_norm": 65.67877197265625,
|
| 141 |
+
"learning_rate": 5.489999999999999e-05,
|
| 142 |
+
"loss": 7.0359,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.010499376599514405,
|
| 147 |
+
"grad_norm": 34.80784225463867,
|
| 148 |
+
"learning_rate": 5.79e-05,
|
| 149 |
+
"loss": 7.0124,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.011024345429490124,
|
| 154 |
+
"grad_norm": 80.05706024169922,
|
| 155 |
+
"learning_rate": 6.0899999999999996e-05,
|
| 156 |
+
"loss": 6.9882,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.011549314259465844,
|
| 161 |
+
"grad_norm": 18.072044372558594,
|
| 162 |
+
"learning_rate": 6.39e-05,
|
| 163 |
+
"loss": 7.082,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.012074283089441565,
|
| 168 |
+
"grad_norm": 27.460533142089844,
|
| 169 |
+
"learning_rate": 6.69e-05,
|
| 170 |
+
"loss": 7.0056,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.012599251919417284,
|
| 175 |
+
"grad_norm": 20.942256927490234,
|
| 176 |
+
"learning_rate": 6.989999999999999e-05,
|
| 177 |
+
"loss": 7.0385,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.013124220749393005,
|
| 182 |
+
"grad_norm": 21.205760955810547,
|
| 183 |
+
"learning_rate": 7.29e-05,
|
| 184 |
+
"loss": 6.9065,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.013649189579368726,
|
| 189 |
+
"grad_norm": 30.357505798339844,
|
| 190 |
+
"learning_rate": 7.589999999999999e-05,
|
| 191 |
+
"loss": 6.9887,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.014174158409344445,
|
| 196 |
+
"grad_norm": 21.6142520904541,
|
| 197 |
+
"learning_rate": 7.89e-05,
|
| 198 |
+
"loss": 7.0671,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.014699127239320165,
|
| 203 |
+
"grad_norm": 34.31865310668945,
|
| 204 |
+
"learning_rate": 8.19e-05,
|
| 205 |
+
"loss": 6.9352,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.015224096069295886,
|
| 210 |
+
"grad_norm": 31.643892288208008,
|
| 211 |
+
"learning_rate": 8.489999999999999e-05,
|
| 212 |
+
"loss": 7.0615,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.015749064899271605,
|
| 217 |
+
"grad_norm": 21.246601104736328,
|
| 218 |
+
"learning_rate": 8.789999999999998e-05,
|
| 219 |
+
"loss": 6.8747,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.016274033729247326,
|
| 224 |
+
"grad_norm": 17.99869155883789,
|
| 225 |
+
"learning_rate": 9.089999999999999e-05,
|
| 226 |
+
"loss": 6.9325,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.016799002559223047,
|
| 231 |
+
"grad_norm": 10.438518524169922,
|
| 232 |
+
"learning_rate": 9.389999999999999e-05,
|
| 233 |
+
"loss": 6.975,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.017323971389198767,
|
| 238 |
+
"grad_norm": 38.324283599853516,
|
| 239 |
+
"learning_rate": 9.69e-05,
|
| 240 |
+
"loss": 6.8833,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.017848940219174488,
|
| 245 |
+
"grad_norm": 25.789045333862305,
|
| 246 |
+
"learning_rate": 9.99e-05,
|
| 247 |
+
"loss": 6.869,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.018373909049150206,
|
| 252 |
+
"grad_norm": 26.649545669555664,
|
| 253 |
+
"learning_rate": 0.0001029,
|
| 254 |
+
"loss": 6.8773,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.018898877879125926,
|
| 259 |
+
"grad_norm": 24.452266693115234,
|
| 260 |
+
"learning_rate": 0.00010589999999999999,
|
| 261 |
+
"loss": 6.7558,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.019423846709101647,
|
| 266 |
+
"grad_norm": 14.234713554382324,
|
| 267 |
+
"learning_rate": 0.00010889999999999999,
|
| 268 |
+
"loss": 6.9624,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.019948815539077368,
|
| 273 |
+
"grad_norm": 16.222957611083984,
|
| 274 |
+
"learning_rate": 0.0001119,
|
| 275 |
+
"loss": 6.9116,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.02047378436905309,
|
| 280 |
+
"grad_norm": 14.955327033996582,
|
| 281 |
+
"learning_rate": 0.00011489999999999999,
|
| 282 |
+
"loss": 6.8227,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.02099875319902881,
|
| 287 |
+
"grad_norm": 13.272610664367676,
|
| 288 |
+
"learning_rate": 0.00011789999999999999,
|
| 289 |
+
"loss": 6.8292,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.021523722029004527,
|
| 294 |
+
"grad_norm": 14.966605186462402,
|
| 295 |
+
"learning_rate": 0.0001209,
|
| 296 |
+
"loss": 6.8364,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.022048690858980247,
|
| 301 |
+
"grad_norm": 11.822840690612793,
|
| 302 |
+
"learning_rate": 0.00012389999999999998,
|
| 303 |
+
"loss": 6.668,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.022573659688955968,
|
| 308 |
+
"grad_norm": 12.50557804107666,
|
| 309 |
+
"learning_rate": 0.0001269,
|
| 310 |
+
"loss": 6.6604,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.02309862851893169,
|
| 315 |
+
"grad_norm": 13.769299507141113,
|
| 316 |
+
"learning_rate": 0.00012989999999999999,
|
| 317 |
+
"loss": 6.6969,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.02362359734890741,
|
| 322 |
+
"grad_norm": 9.62465763092041,
|
| 323 |
+
"learning_rate": 0.00013289999999999998,
|
| 324 |
+
"loss": 6.7391,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.02414856617888313,
|
| 329 |
+
"grad_norm": 8.241267204284668,
|
| 330 |
+
"learning_rate": 0.0001359,
|
| 331 |
+
"loss": 6.7064,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.024673535008858848,
|
| 336 |
+
"grad_norm": 14.394071578979492,
|
| 337 |
+
"learning_rate": 0.0001389,
|
| 338 |
+
"loss": 6.7164,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.02519850383883457,
|
| 343 |
+
"grad_norm": 8.198172569274902,
|
| 344 |
+
"learning_rate": 0.00014189999999999998,
|
| 345 |
+
"loss": 6.7228,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.02572347266881029,
|
| 350 |
+
"grad_norm": 17.51156997680664,
|
| 351 |
+
"learning_rate": 0.00014489999999999997,
|
| 352 |
+
"loss": 6.6933,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.02624844149878601,
|
| 357 |
+
"grad_norm": 14.444514274597168,
|
| 358 |
+
"learning_rate": 0.0001479,
|
| 359 |
+
"loss": 6.5789,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.02677341032876173,
|
| 364 |
+
"grad_norm": 8.710993766784668,
|
| 365 |
+
"learning_rate": 0.00015089999999999998,
|
| 366 |
+
"loss": 6.6053,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.02729837915873745,
|
| 371 |
+
"grad_norm": 11.155731201171875,
|
| 372 |
+
"learning_rate": 0.0001539,
|
| 373 |
+
"loss": 6.6003,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.02782334798871317,
|
| 378 |
+
"grad_norm": 9.0637845993042,
|
| 379 |
+
"learning_rate": 0.0001569,
|
| 380 |
+
"loss": 6.5253,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.02834831681868889,
|
| 385 |
+
"grad_norm": 7.813695907592773,
|
| 386 |
+
"learning_rate": 0.00015989999999999998,
|
| 387 |
+
"loss": 6.5777,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.02887328564866461,
|
| 392 |
+
"grad_norm": 9.640409469604492,
|
| 393 |
+
"learning_rate": 0.0001629,
|
| 394 |
+
"loss": 6.5582,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.02939825447864033,
|
| 399 |
+
"grad_norm": 10.287951469421387,
|
| 400 |
+
"learning_rate": 0.0001659,
|
| 401 |
+
"loss": 6.5429,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.02992322330861605,
|
| 406 |
+
"grad_norm": 8.388657569885254,
|
| 407 |
+
"learning_rate": 0.00016889999999999996,
|
| 408 |
+
"loss": 6.4797,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.030448192138591772,
|
| 413 |
+
"grad_norm": 6.651219844818115,
|
| 414 |
+
"learning_rate": 0.00017189999999999998,
|
| 415 |
+
"loss": 6.472,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.03097316096856749,
|
| 420 |
+
"grad_norm": 6.32115364074707,
|
| 421 |
+
"learning_rate": 0.00017489999999999997,
|
| 422 |
+
"loss": 6.4754,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.03149812979854321,
|
| 427 |
+
"grad_norm": 10.339417457580566,
|
| 428 |
+
"learning_rate": 0.0001779,
|
| 429 |
+
"loss": 6.4067,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.03202309862851893,
|
| 434 |
+
"grad_norm": 7.705801963806152,
|
| 435 |
+
"learning_rate": 0.00018089999999999998,
|
| 436 |
+
"loss": 6.4035,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.03254806745849465,
|
| 441 |
+
"grad_norm": 6.596720218658447,
|
| 442 |
+
"learning_rate": 0.00018389999999999997,
|
| 443 |
+
"loss": 6.3301,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.03307303628847037,
|
| 448 |
+
"grad_norm": 10.351055145263672,
|
| 449 |
+
"learning_rate": 0.0001869,
|
| 450 |
+
"loss": 6.3362,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.03359800511844609,
|
| 455 |
+
"grad_norm": 8.058084487915039,
|
| 456 |
+
"learning_rate": 0.00018989999999999998,
|
| 457 |
+
"loss": 6.4833,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.034122973948421814,
|
| 462 |
+
"grad_norm": 8.515560150146484,
|
| 463 |
+
"learning_rate": 0.0001929,
|
| 464 |
+
"loss": 6.2454,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.034647942778397535,
|
| 469 |
+
"grad_norm": 7.069362640380859,
|
| 470 |
+
"learning_rate": 0.0001959,
|
| 471 |
+
"loss": 6.2539,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.035172911608373256,
|
| 476 |
+
"grad_norm": 6.835115909576416,
|
| 477 |
+
"learning_rate": 0.00019889999999999998,
|
| 478 |
+
"loss": 6.2121,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.035697880438348976,
|
| 483 |
+
"grad_norm": 9.131962776184082,
|
| 484 |
+
"learning_rate": 0.0002019,
|
| 485 |
+
"loss": 6.1163,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.03622284926832469,
|
| 490 |
+
"grad_norm": 6.500721454620361,
|
| 491 |
+
"learning_rate": 0.0002049,
|
| 492 |
+
"loss": 6.2233,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.03674781809830041,
|
| 497 |
+
"grad_norm": 6.955876350402832,
|
| 498 |
+
"learning_rate": 0.00020789999999999996,
|
| 499 |
+
"loss": 6.2746,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.03727278692827613,
|
| 504 |
+
"grad_norm": 12.63397216796875,
|
| 505 |
+
"learning_rate": 0.00021089999999999998,
|
| 506 |
+
"loss": 6.2639,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.03779775575825185,
|
| 511 |
+
"grad_norm": 6.766085624694824,
|
| 512 |
+
"learning_rate": 0.00021389999999999997,
|
| 513 |
+
"loss": 6.2185,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.03832272458822757,
|
| 518 |
+
"grad_norm": 7.125621795654297,
|
| 519 |
+
"learning_rate": 0.0002169,
|
| 520 |
+
"loss": 6.2077,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.038847693418203294,
|
| 525 |
+
"grad_norm": 6.098615646362305,
|
| 526 |
+
"learning_rate": 0.00021989999999999998,
|
| 527 |
+
"loss": 6.1988,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.039372662248179015,
|
| 532 |
+
"grad_norm": 6.724160671234131,
|
| 533 |
+
"learning_rate": 0.00022289999999999997,
|
| 534 |
+
"loss": 6.1645,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.039897631078154736,
|
| 539 |
+
"grad_norm": 6.864727973937988,
|
| 540 |
+
"learning_rate": 0.0002259,
|
| 541 |
+
"loss": 5.9853,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.040422599908130456,
|
| 546 |
+
"grad_norm": 7.888518810272217,
|
| 547 |
+
"learning_rate": 0.00022889999999999998,
|
| 548 |
+
"loss": 6.0459,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.04094756873810618,
|
| 553 |
+
"grad_norm": 5.348147392272949,
|
| 554 |
+
"learning_rate": 0.0002319,
|
| 555 |
+
"loss": 6.0306,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.0414725375680819,
|
| 560 |
+
"grad_norm": 7.666118621826172,
|
| 561 |
+
"learning_rate": 0.0002349,
|
| 562 |
+
"loss": 6.0464,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.04199750639805762,
|
| 567 |
+
"grad_norm": 12.02874755859375,
|
| 568 |
+
"learning_rate": 0.00023789999999999998,
|
| 569 |
+
"loss": 5.9991,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.04252247522803333,
|
| 574 |
+
"grad_norm": 6.495911598205566,
|
| 575 |
+
"learning_rate": 0.0002409,
|
| 576 |
+
"loss": 5.9451,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.04304744405800905,
|
| 581 |
+
"grad_norm": 9.461275100708008,
|
| 582 |
+
"learning_rate": 0.00024389999999999997,
|
| 583 |
+
"loss": 5.9558,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.043572412887984774,
|
| 588 |
+
"grad_norm": 6.616198539733887,
|
| 589 |
+
"learning_rate": 0.0002469,
|
| 590 |
+
"loss": 5.9853,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.044097381717960495,
|
| 595 |
+
"grad_norm": 6.1858086585998535,
|
| 596 |
+
"learning_rate": 0.00024989999999999995,
|
| 597 |
+
"loss": 5.9692,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.044622350547936215,
|
| 602 |
+
"grad_norm": 5.398629188537598,
|
| 603 |
+
"learning_rate": 0.00025289999999999997,
|
| 604 |
+
"loss": 5.9193,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.045147319377911936,
|
| 609 |
+
"grad_norm": 7.980245113372803,
|
| 610 |
+
"learning_rate": 0.0002559,
|
| 611 |
+
"loss": 5.8701,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.04567228820788766,
|
| 616 |
+
"grad_norm": 6.517972946166992,
|
| 617 |
+
"learning_rate": 0.00025889999999999995,
|
| 618 |
+
"loss": 5.905,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.04619725703786338,
|
| 623 |
+
"grad_norm": 6.580096244812012,
|
| 624 |
+
"learning_rate": 0.00026189999999999997,
|
| 625 |
+
"loss": 5.7834,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.0467222258678391,
|
| 630 |
+
"grad_norm": 6.371602535247803,
|
| 631 |
+
"learning_rate": 0.0002649,
|
| 632 |
+
"loss": 5.8452,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.04724719469781482,
|
| 637 |
+
"grad_norm": 6.168027400970459,
|
| 638 |
+
"learning_rate": 0.0002679,
|
| 639 |
+
"loss": 5.6781,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.04777216352779054,
|
| 644 |
+
"grad_norm": 6.000701427459717,
|
| 645 |
+
"learning_rate": 0.00027089999999999997,
|
| 646 |
+
"loss": 5.6675,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.04829713235776626,
|
| 651 |
+
"grad_norm": 6.960968971252441,
|
| 652 |
+
"learning_rate": 0.0002739,
|
| 653 |
+
"loss": 5.5692,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.048822101187741974,
|
| 658 |
+
"grad_norm": 10.075114250183105,
|
| 659 |
+
"learning_rate": 0.0002769,
|
| 660 |
+
"loss": 5.7553,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.049347070017717695,
|
| 665 |
+
"grad_norm": 5.733921527862549,
|
| 666 |
+
"learning_rate": 0.0002799,
|
| 667 |
+
"loss": 5.5125,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.049872038847693416,
|
| 672 |
+
"grad_norm": 7.427346706390381,
|
| 673 |
+
"learning_rate": 0.00028289999999999994,
|
| 674 |
+
"loss": 5.6595,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.05039700767766914,
|
| 679 |
+
"grad_norm": 5.955673694610596,
|
| 680 |
+
"learning_rate": 0.00028589999999999996,
|
| 681 |
+
"loss": 5.5868,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.05092197650764486,
|
| 686 |
+
"grad_norm": 7.7476420402526855,
|
| 687 |
+
"learning_rate": 0.0002889,
|
| 688 |
+
"loss": 5.632,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.05144694533762058,
|
| 693 |
+
"grad_norm": 6.524439811706543,
|
| 694 |
+
"learning_rate": 0.0002919,
|
| 695 |
+
"loss": 5.6333,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.0519719141675963,
|
| 700 |
+
"grad_norm": 6.380852699279785,
|
| 701 |
+
"learning_rate": 0.00029489999999999996,
|
| 702 |
+
"loss": 5.6931,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.05249688299757202,
|
| 707 |
+
"grad_norm": 6.871588706970215,
|
| 708 |
+
"learning_rate": 0.0002979,
|
| 709 |
+
"loss": 5.4705,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.05302185182754774,
|
| 714 |
+
"grad_norm": 5.873289585113525,
|
| 715 |
+
"learning_rate": 0.00029999999999993335,
|
| 716 |
+
"loss": 5.659,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.05354682065752346,
|
| 721 |
+
"grad_norm": 6.722020626068115,
|
| 722 |
+
"learning_rate": 0.00029999999999874875,
|
| 723 |
+
"loss": 5.5263,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.05407178948749918,
|
| 728 |
+
"grad_norm": 7.375940799713135,
|
| 729 |
+
"learning_rate": 0.00029999999999608346,
|
| 730 |
+
"loss": 5.4849,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.0545967583174749,
|
| 735 |
+
"grad_norm": 6.465951919555664,
|
| 736 |
+
"learning_rate": 0.00029999999999193736,
|
| 737 |
+
"loss": 5.4179,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.05512172714745062,
|
| 742 |
+
"grad_norm": 6.1253156661987305,
|
| 743 |
+
"learning_rate": 0.00029999999998631057,
|
| 744 |
+
"loss": 5.415,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.05564669597742634,
|
| 749 |
+
"grad_norm": 6.320364475250244,
|
| 750 |
+
"learning_rate": 0.000299999999979203,
|
| 751 |
+
"loss": 5.2886,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.05617166480740206,
|
| 756 |
+
"grad_norm": 5.7049880027771,
|
| 757 |
+
"learning_rate": 0.00029999999997061473,
|
| 758 |
+
"loss": 5.3008,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.05669663363737778,
|
| 763 |
+
"grad_norm": 7.153939247131348,
|
| 764 |
+
"learning_rate": 0.00029999999996054575,
|
| 765 |
+
"loss": 5.336,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.0572216024673535,
|
| 770 |
+
"grad_norm": 6.331596374511719,
|
| 771 |
+
"learning_rate": 0.00029999999994899596,
|
| 772 |
+
"loss": 5.2242,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.05774657129732922,
|
| 777 |
+
"grad_norm": 6.270790100097656,
|
| 778 |
+
"learning_rate": 0.00029999999993596553,
|
| 779 |
+
"loss": 5.2873,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.05827154012730494,
|
| 784 |
+
"grad_norm": 6.87438440322876,
|
| 785 |
+
"learning_rate": 0.0002999999999214543,
|
| 786 |
+
"loss": 5.2111,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.05879650895728066,
|
| 791 |
+
"grad_norm": 6.366969585418701,
|
| 792 |
+
"learning_rate": 0.0002999999999054623,
|
| 793 |
+
"loss": 5.1685,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.05932147778725638,
|
| 798 |
+
"grad_norm": 6.2817816734313965,
|
| 799 |
+
"learning_rate": 0.00029999999988798964,
|
| 800 |
+
"loss": 5.3457,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.0598464466172321,
|
| 805 |
+
"grad_norm": 5.789464950561523,
|
| 806 |
+
"learning_rate": 0.0002999999998690362,
|
| 807 |
+
"loss": 5.0926,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.060371415447207824,
|
| 812 |
+
"grad_norm": 6.459042072296143,
|
| 813 |
+
"learning_rate": 0.00029999999984860204,
|
| 814 |
+
"loss": 5.0385,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.060896384277183545,
|
| 819 |
+
"grad_norm": 6.7204413414001465,
|
| 820 |
+
"learning_rate": 0.00029999999982668717,
|
| 821 |
+
"loss": 5.2199,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.061421353107159266,
|
| 826 |
+
"grad_norm": 6.44543981552124,
|
| 827 |
+
"learning_rate": 0.0002999999998032915,
|
| 828 |
+
"loss": 5.1449,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.06194632193713498,
|
| 833 |
+
"grad_norm": 5.885377407073975,
|
| 834 |
+
"learning_rate": 0.0002999999997784151,
|
| 835 |
+
"loss": 5.1175,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.0624712907671107,
|
| 840 |
+
"grad_norm": 6.293627738952637,
|
| 841 |
+
"learning_rate": 0.000299999999752058,
|
| 842 |
+
"loss": 5.0933,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.06299625959708642,
|
| 847 |
+
"grad_norm": 6.686017990112305,
|
| 848 |
+
"learning_rate": 0.0002999999997242202,
|
| 849 |
+
"loss": 4.9613,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.06352122842706215,
|
| 854 |
+
"grad_norm": 5.605543613433838,
|
| 855 |
+
"learning_rate": 0.0002999999996949016,
|
| 856 |
+
"loss": 5.0621,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.06404619725703786,
|
| 861 |
+
"grad_norm": 6.000121593475342,
|
| 862 |
+
"learning_rate": 0.0002999999996641022,
|
| 863 |
+
"loss": 5.0555,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.06457116608701359,
|
| 868 |
+
"grad_norm": 7.083418369293213,
|
| 869 |
+
"learning_rate": 0.00029999999963182215,
|
| 870 |
+
"loss": 5.0052,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.0650961349169893,
|
| 875 |
+
"grad_norm": 6.544703483581543,
|
| 876 |
+
"learning_rate": 0.00029999999959806135,
|
| 877 |
+
"loss": 5.0612,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.06562110374696502,
|
| 882 |
+
"grad_norm": 6.990364074707031,
|
| 883 |
+
"learning_rate": 0.0002999999995628198,
|
| 884 |
+
"loss": 5.0725,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.06614607257694075,
|
| 889 |
+
"grad_norm": 6.532026290893555,
|
| 890 |
+
"learning_rate": 0.0002999999995260976,
|
| 891 |
+
"loss": 5.0609,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.06667104140691646,
|
| 896 |
+
"grad_norm": 6.3640522956848145,
|
| 897 |
+
"learning_rate": 0.00029999999948789453,
|
| 898 |
+
"loss": 4.9563,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.06719601023689219,
|
| 903 |
+
"grad_norm": 7.006086349487305,
|
| 904 |
+
"learning_rate": 0.00029999999944821084,
|
| 905 |
+
"loss": 4.9507,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.0677209790668679,
|
| 910 |
+
"grad_norm": 6.476901531219482,
|
| 911 |
+
"learning_rate": 0.00029999999940704634,
|
| 912 |
+
"loss": 4.9325,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.06824594789684363,
|
| 917 |
+
"grad_norm": 6.32811975479126,
|
| 918 |
+
"learning_rate": 0.00029999999936440114,
|
| 919 |
+
"loss": 4.819,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.06877091672681934,
|
| 924 |
+
"grad_norm": 5.81315279006958,
|
| 925 |
+
"learning_rate": 0.00029999999932027514,
|
| 926 |
+
"loss": 4.8838,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.06929588555679507,
|
| 931 |
+
"grad_norm": 6.884243488311768,
|
| 932 |
+
"learning_rate": 0.00029999999927466845,
|
| 933 |
+
"loss": 4.9064,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.06982085438677078,
|
| 938 |
+
"grad_norm": 5.8015313148498535,
|
| 939 |
+
"learning_rate": 0.00029999999922758106,
|
| 940 |
+
"loss": 4.8818,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.07034582321674651,
|
| 945 |
+
"grad_norm": 6.7853875160217285,
|
| 946 |
+
"learning_rate": 0.00029999999917901287,
|
| 947 |
+
"loss": 4.7985,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.07087079204672223,
|
| 952 |
+
"grad_norm": 6.297364711761475,
|
| 953 |
+
"learning_rate": 0.000299999999128964,
|
| 954 |
+
"loss": 4.8261,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.07139576087669795,
|
| 959 |
+
"grad_norm": 7.7735514640808105,
|
| 960 |
+
"learning_rate": 0.0002999999990774343,
|
| 961 |
+
"loss": 4.8545,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.07192072970667367,
|
| 966 |
+
"grad_norm": 6.501247406005859,
|
| 967 |
+
"learning_rate": 0.00029999999902442396,
|
| 968 |
+
"loss": 4.7666,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.07244569853664938,
|
| 973 |
+
"grad_norm": 5.721576690673828,
|
| 974 |
+
"learning_rate": 0.00029999999896993283,
|
| 975 |
+
"loss": 4.6484,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.07297066736662511,
|
| 980 |
+
"grad_norm": 6.3349127769470215,
|
| 981 |
+
"learning_rate": 0.000299999998913961,
|
| 982 |
+
"loss": 4.6921,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.07349563619660082,
|
| 987 |
+
"grad_norm": 5.776736259460449,
|
| 988 |
+
"learning_rate": 0.0002999999988565084,
|
| 989 |
+
"loss": 4.6327,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.07402060502657655,
|
| 994 |
+
"grad_norm": 8.254961013793945,
|
| 995 |
+
"learning_rate": 0.0002999999987975751,
|
| 996 |
+
"loss": 4.7353,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.07454557385655226,
|
| 1001 |
+
"grad_norm": 7.920860767364502,
|
| 1002 |
+
"learning_rate": 0.000299999998737161,
|
| 1003 |
+
"loss": 4.6561,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.07507054268652799,
|
| 1008 |
+
"grad_norm": 6.328056812286377,
|
| 1009 |
+
"learning_rate": 0.0002999999986752662,
|
| 1010 |
+
"loss": 4.556,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.0755955115165037,
|
| 1015 |
+
"grad_norm": 6.354831695556641,
|
| 1016 |
+
"learning_rate": 0.0002999999986118907,
|
| 1017 |
+
"loss": 4.5249,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.07612048034647943,
|
| 1022 |
+
"grad_norm": 5.605742931365967,
|
| 1023 |
+
"learning_rate": 0.0002999999985470344,
|
| 1024 |
+
"loss": 4.5733,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.07664544917645515,
|
| 1029 |
+
"grad_norm": 7.2531514167785645,
|
| 1030 |
+
"learning_rate": 0.0002999999984806974,
|
| 1031 |
+
"loss": 4.6005,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.07717041800643087,
|
| 1036 |
+
"grad_norm": 6.168049335479736,
|
| 1037 |
+
"learning_rate": 0.0002999999984128797,
|
| 1038 |
+
"loss": 4.4608,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.07769538683640659,
|
| 1043 |
+
"grad_norm": 5.962798595428467,
|
| 1044 |
+
"learning_rate": 0.00029999999834358116,
|
| 1045 |
+
"loss": 4.4705,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.0782203556663823,
|
| 1050 |
+
"grad_norm": 7.6847357749938965,
|
| 1051 |
+
"learning_rate": 0.00029999999827280195,
|
| 1052 |
+
"loss": 4.5147,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.07874532449635803,
|
| 1057 |
+
"grad_norm": 8.098923683166504,
|
| 1058 |
+
"learning_rate": 0.000299999998200542,
|
| 1059 |
+
"loss": 4.5436,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.07927029332633374,
|
| 1064 |
+
"grad_norm": 6.158365249633789,
|
| 1065 |
+
"learning_rate": 0.00029999999812680133,
|
| 1066 |
+
"loss": 4.4493,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.07979526215630947,
|
| 1071 |
+
"grad_norm": 6.325509548187256,
|
| 1072 |
+
"learning_rate": 0.00029999999805157987,
|
| 1073 |
+
"loss": 4.4293,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.08032023098628518,
|
| 1078 |
+
"grad_norm": 6.080830097198486,
|
| 1079 |
+
"learning_rate": 0.0002999999979748777,
|
| 1080 |
+
"loss": 4.3919,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.08084519981626091,
|
| 1085 |
+
"grad_norm": 7.090545654296875,
|
| 1086 |
+
"learning_rate": 0.0002999999978966948,
|
| 1087 |
+
"loss": 4.4191,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.08137016864623663,
|
| 1092 |
+
"grad_norm": 6.284291744232178,
|
| 1093 |
+
"learning_rate": 0.00029999999781703116,
|
| 1094 |
+
"loss": 4.3456,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.08189513747621235,
|
| 1099 |
+
"grad_norm": 6.383409023284912,
|
| 1100 |
+
"learning_rate": 0.0002999999977358868,
|
| 1101 |
+
"loss": 4.3264,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.08242010630618807,
|
| 1106 |
+
"grad_norm": 6.384091854095459,
|
| 1107 |
+
"learning_rate": 0.00029999999765326166,
|
| 1108 |
+
"loss": 4.265,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.0829450751361638,
|
| 1113 |
+
"grad_norm": 6.6042022705078125,
|
| 1114 |
+
"learning_rate": 0.0002999999975691559,
|
| 1115 |
+
"loss": 4.4527,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.08347004396613951,
|
| 1120 |
+
"grad_norm": 5.603631496429443,
|
| 1121 |
+
"learning_rate": 0.0002999999974835692,
|
| 1122 |
+
"loss": 4.3473,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.08399501279611524,
|
| 1127 |
+
"grad_norm": 5.590120315551758,
|
| 1128 |
+
"learning_rate": 0.00029999999739650194,
|
| 1129 |
+
"loss": 4.2386,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.08451998162609095,
|
| 1134 |
+
"grad_norm": 6.289425849914551,
|
| 1135 |
+
"learning_rate": 0.00029999999730795384,
|
| 1136 |
+
"loss": 4.2665,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.08504495045606666,
|
| 1141 |
+
"grad_norm": 6.349957466125488,
|
| 1142 |
+
"learning_rate": 0.00029999999721792506,
|
| 1143 |
+
"loss": 4.2492,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.08556991928604239,
|
| 1148 |
+
"grad_norm": 6.4022722244262695,
|
| 1149 |
+
"learning_rate": 0.0002999999971264155,
|
| 1150 |
+
"loss": 4.2056,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.0860948881160181,
|
| 1155 |
+
"grad_norm": 7.056876182556152,
|
| 1156 |
+
"learning_rate": 0.00029999999703342523,
|
| 1157 |
+
"loss": 4.1929,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.08661985694599383,
|
| 1162 |
+
"grad_norm": 6.141146659851074,
|
| 1163 |
+
"learning_rate": 0.00029999999693895425,
|
| 1164 |
+
"loss": 4.2077,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.08714482577596955,
|
| 1169 |
+
"grad_norm": 6.444318771362305,
|
| 1170 |
+
"learning_rate": 0.0002999999968430025,
|
| 1171 |
+
"loss": 4.1848,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.08766979460594528,
|
| 1176 |
+
"grad_norm": 6.252465724945068,
|
| 1177 |
+
"learning_rate": 0.00029999999674557005,
|
| 1178 |
+
"loss": 4.1801,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.08819476343592099,
|
| 1183 |
+
"grad_norm": 6.427366256713867,
|
| 1184 |
+
"learning_rate": 0.0002999999966466568,
|
| 1185 |
+
"loss": 4.2509,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.08871973226589672,
|
| 1190 |
+
"grad_norm": 6.234042167663574,
|
| 1191 |
+
"learning_rate": 0.0002999999965462629,
|
| 1192 |
+
"loss": 4.1133,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.08924470109587243,
|
| 1197 |
+
"grad_norm": 6.525606155395508,
|
| 1198 |
+
"learning_rate": 0.00029999999644438817,
|
| 1199 |
+
"loss": 4.019,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.0005249688299757202,
|
| 1204 |
+
"grad_norm": 5.570987701416016,
|
| 1205 |
+
"learning_rate": 0.00029999999634103275,
|
| 1206 |
+
"loss": 4.1462,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.0010499376599514404,
|
| 1211 |
+
"grad_norm": 5.909586429595947,
|
| 1212 |
+
"learning_rate": 0.0002999999962361966,
|
| 1213 |
+
"loss": 4.126,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.0015749064899271605,
|
| 1218 |
+
"grad_norm": 7.547698020935059,
|
| 1219 |
+
"learning_rate": 0.00029999999612987966,
|
| 1220 |
+
"loss": 4.0706,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.002099875319902881,
|
| 1225 |
+
"grad_norm": 6.932499408721924,
|
| 1226 |
+
"learning_rate": 0.00029999999602208204,
|
| 1227 |
+
"loss": 4.1576,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.002624844149878601,
|
| 1232 |
+
"grad_norm": 7.4274797439575195,
|
| 1233 |
+
"learning_rate": 0.0002999999959128037,
|
| 1234 |
+
"loss": 4.1343,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.003149812979854321,
|
| 1239 |
+
"grad_norm": 6.038876056671143,
|
| 1240 |
+
"learning_rate": 0.00029999999580204457,
|
| 1241 |
+
"loss": 4.1781,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.0036747818098300414,
|
| 1246 |
+
"grad_norm": 6.615757465362549,
|
| 1247 |
+
"learning_rate": 0.0002999999956898047,
|
| 1248 |
+
"loss": 4.0667,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.004199750639805762,
|
| 1253 |
+
"grad_norm": 6.53090238571167,
|
| 1254 |
+
"learning_rate": 0.00029999999557608415,
|
| 1255 |
+
"loss": 4.1289,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.0047247194697814816,
|
| 1260 |
+
"grad_norm": 5.512113571166992,
|
| 1261 |
+
"learning_rate": 0.0002999999954608828,
|
| 1262 |
+
"loss": 3.9358,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.005249688299757202,
|
| 1267 |
+
"grad_norm": 6.99248743057251,
|
| 1268 |
+
"learning_rate": 0.00029999999534420074,
|
| 1269 |
+
"loss": 3.9963,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.005774657129732922,
|
| 1274 |
+
"grad_norm": 5.911637306213379,
|
| 1275 |
+
"learning_rate": 0.00029999999522603793,
|
| 1276 |
+
"loss": 4.017,
|
| 1277 |
+
"step": 1810
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.006299625959708642,
|
| 1281 |
+
"grad_norm": 5.9147562980651855,
|
| 1282 |
+
"learning_rate": 0.00029999999510639443,
|
| 1283 |
+
"loss": 4.0137,
|
| 1284 |
+
"step": 1820
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 0.006824594789684363,
|
| 1288 |
+
"grad_norm": 5.845144271850586,
|
| 1289 |
+
"learning_rate": 0.00029999999498527013,
|
| 1290 |
+
"loss": 4.0678,
|
| 1291 |
+
"step": 1830
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 0.007349563619660083,
|
| 1295 |
+
"grad_norm": 5.957810401916504,
|
| 1296 |
+
"learning_rate": 0.00029999999486266513,
|
| 1297 |
+
"loss": 3.891,
|
| 1298 |
+
"step": 1840
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.007874532449635803,
|
| 1302 |
+
"grad_norm": 6.678396224975586,
|
| 1303 |
+
"learning_rate": 0.0002999999947385794,
|
| 1304 |
+
"loss": 3.9104,
|
| 1305 |
+
"step": 1850
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.008399501279611523,
|
| 1309 |
+
"grad_norm": 5.920273303985596,
|
| 1310 |
+
"learning_rate": 0.0002999999946130129,
|
| 1311 |
+
"loss": 3.9749,
|
| 1312 |
+
"step": 1860
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 0.008924470109587244,
|
| 1316 |
+
"grad_norm": 6.2641777992248535,
|
| 1317 |
+
"learning_rate": 0.0002999999944859657,
|
| 1318 |
+
"loss": 3.8296,
|
| 1319 |
+
"step": 1870
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 0.009449438939562963,
|
| 1323 |
+
"grad_norm": 6.42849588394165,
|
| 1324 |
+
"learning_rate": 0.00029999999435743776,
|
| 1325 |
+
"loss": 3.9217,
|
| 1326 |
+
"step": 1880
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.009974407769538684,
|
| 1330 |
+
"grad_norm": 5.971102714538574,
|
| 1331 |
+
"learning_rate": 0.0002999999942274291,
|
| 1332 |
+
"loss": 3.8751,
|
| 1333 |
+
"step": 1890
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 0.010499376599514405,
|
| 1337 |
+
"grad_norm": 6.6937408447265625,
|
| 1338 |
+
"learning_rate": 0.00029999999409593964,
|
| 1339 |
+
"loss": 3.8184,
|
| 1340 |
+
"step": 1900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.011024345429490124,
|
| 1344 |
+
"grad_norm": 6.671769618988037,
|
| 1345 |
+
"learning_rate": 0.0002999999939629695,
|
| 1346 |
+
"loss": 3.7852,
|
| 1347 |
+
"step": 1910
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 0.011549314259465844,
|
| 1351 |
+
"grad_norm": 6.817453861236572,
|
| 1352 |
+
"learning_rate": 0.00029999999382851857,
|
| 1353 |
+
"loss": 3.8804,
|
| 1354 |
+
"step": 1920
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.012074283089441565,
|
| 1358 |
+
"grad_norm": 7.681510925292969,
|
| 1359 |
+
"learning_rate": 0.00029999999369258694,
|
| 1360 |
+
"loss": 3.8054,
|
| 1361 |
+
"step": 1930
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 0.012599251919417284,
|
| 1365 |
+
"grad_norm": 6.021421432495117,
|
| 1366 |
+
"learning_rate": 0.00029999999355517456,
|
| 1367 |
+
"loss": 3.8191,
|
| 1368 |
+
"step": 1940
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 0.013124220749393005,
|
| 1372 |
+
"grad_norm": 6.368685245513916,
|
| 1373 |
+
"learning_rate": 0.0002999999934162815,
|
| 1374 |
+
"loss": 3.671,
|
| 1375 |
+
"step": 1950
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 0.013649189579368726,
|
| 1379 |
+
"grad_norm": 6.735239505767822,
|
| 1380 |
+
"learning_rate": 0.0002999999932759076,
|
| 1381 |
+
"loss": 3.8424,
|
| 1382 |
+
"step": 1960
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.014174158409344445,
|
| 1386 |
+
"grad_norm": 6.516197681427002,
|
| 1387 |
+
"learning_rate": 0.00029999999313405304,
|
| 1388 |
+
"loss": 3.9186,
|
| 1389 |
+
"step": 1970
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 0.014699127239320165,
|
| 1393 |
+
"grad_norm": 6.4768476486206055,
|
| 1394 |
+
"learning_rate": 0.0002999999929907177,
|
| 1395 |
+
"loss": 3.7679,
|
| 1396 |
+
"step": 1980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 0.015224096069295886,
|
| 1400 |
+
"grad_norm": 6.700578212738037,
|
| 1401 |
+
"learning_rate": 0.00029999999284590164,
|
| 1402 |
+
"loss": 3.904,
|
| 1403 |
+
"step": 1990
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 0.015749064899271605,
|
| 1407 |
+
"grad_norm": 6.943190097808838,
|
| 1408 |
+
"learning_rate": 0.0002999999926996049,
|
| 1409 |
+
"loss": 3.7242,
|
| 1410 |
+
"step": 2000
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.016274033729247326,
|
| 1414 |
+
"grad_norm": 5.892586708068848,
|
| 1415 |
+
"learning_rate": 0.0002999999925518273,
|
| 1416 |
+
"loss": 3.7425,
|
| 1417 |
+
"step": 2010
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.016799002559223047,
|
| 1421 |
+
"grad_norm": 6.512500286102295,
|
| 1422 |
+
"learning_rate": 0.0002999999924025691,
|
| 1423 |
+
"loss": 3.797,
|
| 1424 |
+
"step": 2020
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.017323971389198767,
|
| 1428 |
+
"grad_norm": 6.24969482421875,
|
| 1429 |
+
"learning_rate": 0.0002999999922518301,
|
| 1430 |
+
"loss": 3.7444,
|
| 1431 |
+
"step": 2030
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.017848940219174488,
|
| 1435 |
+
"grad_norm": 6.845114707946777,
|
| 1436 |
+
"learning_rate": 0.0002999999920996103,
|
| 1437 |
+
"loss": 3.6969,
|
| 1438 |
+
"step": 2040
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 0.018373909049150206,
|
| 1442 |
+
"grad_norm": 6.944314002990723,
|
| 1443 |
+
"learning_rate": 0.00029999999194590986,
|
| 1444 |
+
"loss": 3.731,
|
| 1445 |
+
"step": 2050
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 0.018898877879125926,
|
| 1449 |
+
"grad_norm": 6.697751998901367,
|
| 1450 |
+
"learning_rate": 0.00029999999179072866,
|
| 1451 |
+
"loss": 3.5394,
|
| 1452 |
+
"step": 2060
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 0.019423846709101647,
|
| 1456 |
+
"grad_norm": 6.382928848266602,
|
| 1457 |
+
"learning_rate": 0.0002999999916340667,
|
| 1458 |
+
"loss": 3.8336,
|
| 1459 |
+
"step": 2070
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 0.019948815539077368,
|
| 1463 |
+
"grad_norm": 6.065791130065918,
|
| 1464 |
+
"learning_rate": 0.000299999991475924,
|
| 1465 |
+
"loss": 3.746,
|
| 1466 |
+
"step": 2080
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.02047378436905309,
|
| 1470 |
+
"grad_norm": 6.519564151763916,
|
| 1471 |
+
"learning_rate": 0.0002999999913163006,
|
| 1472 |
+
"loss": 3.6768,
|
| 1473 |
+
"step": 2090
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 0.02099875319902881,
|
| 1477 |
+
"grad_norm": 6.374921798706055,
|
| 1478 |
+
"learning_rate": 0.0002999999911551964,
|
| 1479 |
+
"loss": 3.6625,
|
| 1480 |
+
"step": 2100
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 0.021523722029004527,
|
| 1484 |
+
"grad_norm": 6.67402982711792,
|
| 1485 |
+
"learning_rate": 0.00029999999099261155,
|
| 1486 |
+
"loss": 3.7065,
|
| 1487 |
+
"step": 2110
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 0.022048690858980247,
|
| 1491 |
+
"grad_norm": 6.103598594665527,
|
| 1492 |
+
"learning_rate": 0.0002999999908285459,
|
| 1493 |
+
"loss": 3.5128,
|
| 1494 |
+
"step": 2120
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 0.022573659688955968,
|
| 1498 |
+
"grad_norm": 6.75986909866333,
|
| 1499 |
+
"learning_rate": 0.0002999999906629995,
|
| 1500 |
+
"loss": 3.4612,
|
| 1501 |
+
"step": 2130
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 0.02309862851893169,
|
| 1505 |
+
"grad_norm": 6.420792102813721,
|
| 1506 |
+
"learning_rate": 0.0002999999904959724,
|
| 1507 |
+
"loss": 3.5584,
|
| 1508 |
+
"step": 2140
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.02362359734890741,
|
| 1512 |
+
"grad_norm": 8.478610038757324,
|
| 1513 |
+
"learning_rate": 0.0002999999903274646,
|
| 1514 |
+
"loss": 3.6476,
|
| 1515 |
+
"step": 2150
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 0.02414856617888313,
|
| 1519 |
+
"grad_norm": 7.021547317504883,
|
| 1520 |
+
"learning_rate": 0.000299999990157476,
|
| 1521 |
+
"loss": 3.5927,
|
| 1522 |
+
"step": 2160
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.024673535008858848,
|
| 1526 |
+
"grad_norm": 6.317281723022461,
|
| 1527 |
+
"learning_rate": 0.00029999998998600665,
|
| 1528 |
+
"loss": 3.6241,
|
| 1529 |
+
"step": 2170
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 0.02519850383883457,
|
| 1533 |
+
"grad_norm": 5.901510715484619,
|
| 1534 |
+
"learning_rate": 0.0002999999898130566,
|
| 1535 |
+
"loss": 3.6491,
|
| 1536 |
+
"step": 2180
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 0.02572347266881029,
|
| 1540 |
+
"grad_norm": 6.88697624206543,
|
| 1541 |
+
"learning_rate": 0.00029999998963862583,
|
| 1542 |
+
"loss": 3.619,
|
| 1543 |
+
"step": 2190
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 0.02624844149878601,
|
| 1547 |
+
"grad_norm": 6.949872016906738,
|
| 1548 |
+
"learning_rate": 0.0002999999894627143,
|
| 1549 |
+
"loss": 3.4568,
|
| 1550 |
+
"step": 2200
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.02677341032876173,
|
| 1554 |
+
"grad_norm": 6.634612560272217,
|
| 1555 |
+
"learning_rate": 0.000299999989285322,
|
| 1556 |
+
"loss": 3.5583,
|
| 1557 |
+
"step": 2210
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 0.02729837915873745,
|
| 1561 |
+
"grad_norm": 8.479588508605957,
|
| 1562 |
+
"learning_rate": 0.00029999998910644905,
|
| 1563 |
+
"loss": 3.5644,
|
| 1564 |
+
"step": 2220
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 0.02782334798871317,
|
| 1568 |
+
"grad_norm": 6.708542346954346,
|
| 1569 |
+
"learning_rate": 0.0002999999889260953,
|
| 1570 |
+
"loss": 3.4792,
|
| 1571 |
+
"step": 2230
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.02834831681868889,
|
| 1575 |
+
"grad_norm": 7.055044174194336,
|
| 1576 |
+
"learning_rate": 0.00029999998874426085,
|
| 1577 |
+
"loss": 3.5505,
|
| 1578 |
+
"step": 2240
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.02887328564866461,
|
| 1582 |
+
"grad_norm": 6.504602432250977,
|
| 1583 |
+
"learning_rate": 0.0002999999885609456,
|
| 1584 |
+
"loss": 3.5346,
|
| 1585 |
+
"step": 2250
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.02939825447864033,
|
| 1589 |
+
"grad_norm": 6.789810657501221,
|
| 1590 |
+
"learning_rate": 0.0002999999883761497,
|
| 1591 |
+
"loss": 3.5312,
|
| 1592 |
+
"step": 2260
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.02992322330861605,
|
| 1596 |
+
"grad_norm": 6.327883720397949,
|
| 1597 |
+
"learning_rate": 0.000299999988189873,
|
| 1598 |
+
"loss": 3.4453,
|
| 1599 |
+
"step": 2270
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 0.030448192138591772,
|
| 1603 |
+
"grad_norm": 6.850128173828125,
|
| 1604 |
+
"learning_rate": 0.00029999998800211557,
|
| 1605 |
+
"loss": 3.4946,
|
| 1606 |
+
"step": 2280
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 0.03097316096856749,
|
| 1610 |
+
"grad_norm": 6.40550422668457,
|
| 1611 |
+
"learning_rate": 0.0002999999878128774,
|
| 1612 |
+
"loss": 3.4903,
|
| 1613 |
+
"step": 2290
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 0.03149812979854321,
|
| 1617 |
+
"grad_norm": 6.313216209411621,
|
| 1618 |
+
"learning_rate": 0.0002999999876221585,
|
| 1619 |
+
"loss": 3.4033,
|
| 1620 |
+
"step": 2300
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 0.03202309862851893,
|
| 1624 |
+
"grad_norm": 7.214779376983643,
|
| 1625 |
+
"learning_rate": 0.0002999999874299589,
|
| 1626 |
+
"loss": 3.4198,
|
| 1627 |
+
"step": 2310
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 0.03254806745849465,
|
| 1631 |
+
"grad_norm": 6.715667247772217,
|
| 1632 |
+
"learning_rate": 0.00029999998723627847,
|
| 1633 |
+
"loss": 3.3777,
|
| 1634 |
+
"step": 2320
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.03307303628847037,
|
| 1638 |
+
"grad_norm": 5.828542709350586,
|
| 1639 |
+
"learning_rate": 0.0002999999870411174,
|
| 1640 |
+
"loss": 3.405,
|
| 1641 |
+
"step": 2330
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 0.03359800511844609,
|
| 1645 |
+
"grad_norm": 7.116888999938965,
|
| 1646 |
+
"learning_rate": 0.00029999998684447556,
|
| 1647 |
+
"loss": 3.6236,
|
| 1648 |
+
"step": 2340
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 0.034122973948421814,
|
| 1652 |
+
"grad_norm": 6.714303016662598,
|
| 1653 |
+
"learning_rate": 0.000299999986646353,
|
| 1654 |
+
"loss": 3.2953,
|
| 1655 |
+
"step": 2350
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 0.034647942778397535,
|
| 1659 |
+
"grad_norm": 6.0912089347839355,
|
| 1660 |
+
"learning_rate": 0.0002999999864467497,
|
| 1661 |
+
"loss": 3.3727,
|
| 1662 |
+
"step": 2360
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 0.035172911608373256,
|
| 1666 |
+
"grad_norm": 6.161454677581787,
|
| 1667 |
+
"learning_rate": 0.00029999998624566566,
|
| 1668 |
+
"loss": 3.2791,
|
| 1669 |
+
"step": 2370
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.035697880438348976,
|
| 1673 |
+
"grad_norm": 6.854067802429199,
|
| 1674 |
+
"learning_rate": 0.00029999998604310086,
|
| 1675 |
+
"loss": 3.2219,
|
| 1676 |
+
"step": 2380
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.03622284926832469,
|
| 1680 |
+
"grad_norm": 6.076297760009766,
|
| 1681 |
+
"learning_rate": 0.0002999999858390553,
|
| 1682 |
+
"loss": 3.3636,
|
| 1683 |
+
"step": 2390
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.03674781809830041,
|
| 1687 |
+
"grad_norm": 6.223069190979004,
|
| 1688 |
+
"learning_rate": 0.00029999998563352907,
|
| 1689 |
+
"loss": 3.457,
|
| 1690 |
+
"step": 2400
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.03727278692827613,
|
| 1694 |
+
"grad_norm": 6.591337203979492,
|
| 1695 |
+
"learning_rate": 0.0002999999854265221,
|
| 1696 |
+
"loss": 3.4884,
|
| 1697 |
+
"step": 2410
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.03779775575825185,
|
| 1701 |
+
"grad_norm": 9.24142837524414,
|
| 1702 |
+
"learning_rate": 0.00029999998521803434,
|
| 1703 |
+
"loss": 3.419,
|
| 1704 |
+
"step": 2420
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.03832272458822757,
|
| 1708 |
+
"grad_norm": 7.019595146179199,
|
| 1709 |
+
"learning_rate": 0.0002999999850080659,
|
| 1710 |
+
"loss": 3.427,
|
| 1711 |
+
"step": 2430
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.038847693418203294,
|
| 1715 |
+
"grad_norm": 6.758765697479248,
|
| 1716 |
+
"learning_rate": 0.00029999998479661666,
|
| 1717 |
+
"loss": 3.4453,
|
| 1718 |
+
"step": 2440
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.039372662248179015,
|
| 1722 |
+
"grad_norm": 7.150947570800781,
|
| 1723 |
+
"learning_rate": 0.00029999998458368673,
|
| 1724 |
+
"loss": 3.4291,
|
| 1725 |
+
"step": 2450
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.039897631078154736,
|
| 1729 |
+
"grad_norm": 6.2172136306762695,
|
| 1730 |
+
"learning_rate": 0.00029999998436927604,
|
| 1731 |
+
"loss": 3.1965,
|
| 1732 |
+
"step": 2460
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.040422599908130456,
|
| 1736 |
+
"grad_norm": 6.834853649139404,
|
| 1737 |
+
"learning_rate": 0.0002999999841533846,
|
| 1738 |
+
"loss": 3.3453,
|
| 1739 |
+
"step": 2470
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.04094756873810618,
|
| 1743 |
+
"grad_norm": 7.9789137840271,
|
| 1744 |
+
"learning_rate": 0.00029999998393601243,
|
| 1745 |
+
"loss": 3.3349,
|
| 1746 |
+
"step": 2480
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.0414725375680819,
|
| 1750 |
+
"grad_norm": 6.828987121582031,
|
| 1751 |
+
"learning_rate": 0.00029999998371715955,
|
| 1752 |
+
"loss": 3.393,
|
| 1753 |
+
"step": 2490
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 0.04199750639805762,
|
| 1757 |
+
"grad_norm": 7.442930221557617,
|
| 1758 |
+
"learning_rate": 0.00029999998349682593,
|
| 1759 |
+
"loss": 3.3438,
|
| 1760 |
+
"step": 2500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.04252247522803333,
|
| 1764 |
+
"grad_norm": 7.22468376159668,
|
| 1765 |
+
"learning_rate": 0.00029999998327501155,
|
| 1766 |
+
"loss": 3.3022,
|
| 1767 |
+
"step": 2510
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 0.04304744405800905,
|
| 1771 |
+
"grad_norm": 7.090549468994141,
|
| 1772 |
+
"learning_rate": 0.0002999999830517165,
|
| 1773 |
+
"loss": 3.3002,
|
| 1774 |
+
"step": 2520
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 0.043572412887984774,
|
| 1778 |
+
"grad_norm": 7.341608047485352,
|
| 1779 |
+
"learning_rate": 0.00029999998282694067,
|
| 1780 |
+
"loss": 3.351,
|
| 1781 |
+
"step": 2530
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 0.044097381717960495,
|
| 1785 |
+
"grad_norm": 6.907212257385254,
|
| 1786 |
+
"learning_rate": 0.00029999998260068404,
|
| 1787 |
+
"loss": 3.3846,
|
| 1788 |
+
"step": 2540
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 0.044622350547936215,
|
| 1792 |
+
"grad_norm": 6.958861351013184,
|
| 1793 |
+
"learning_rate": 0.0002999999823729468,
|
| 1794 |
+
"loss": 3.3186,
|
| 1795 |
+
"step": 2550
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 0.045147319377911936,
|
| 1799 |
+
"grad_norm": 7.10768461227417,
|
| 1800 |
+
"learning_rate": 0.0002999999821437287,
|
| 1801 |
+
"loss": 3.3192,
|
| 1802 |
+
"step": 2560
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.04567228820788766,
|
| 1806 |
+
"grad_norm": 6.437366962432861,
|
| 1807 |
+
"learning_rate": 0.0002999999819130299,
|
| 1808 |
+
"loss": 3.3617,
|
| 1809 |
+
"step": 2570
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 0.04619725703786338,
|
| 1813 |
+
"grad_norm": 7.021583557128906,
|
| 1814 |
+
"learning_rate": 0.0002999999816808504,
|
| 1815 |
+
"loss": 3.2424,
|
| 1816 |
+
"step": 2580
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 0.0467222258678391,
|
| 1820 |
+
"grad_norm": 7.078067302703857,
|
| 1821 |
+
"learning_rate": 0.00029999998144719013,
|
| 1822 |
+
"loss": 3.3379,
|
| 1823 |
+
"step": 2590
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 0.04724719469781482,
|
| 1827 |
+
"grad_norm": 6.58331823348999,
|
| 1828 |
+
"learning_rate": 0.0002999999812120492,
|
| 1829 |
+
"loss": 3.1476,
|
| 1830 |
+
"step": 2600
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 0.04777216352779054,
|
| 1834 |
+
"grad_norm": 7.511198997497559,
|
| 1835 |
+
"learning_rate": 0.0002999999809754274,
|
| 1836 |
+
"loss": 3.1552,
|
| 1837 |
+
"step": 2610
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 0.04829713235776626,
|
| 1841 |
+
"grad_norm": 6.889819622039795,
|
| 1842 |
+
"learning_rate": 0.00029999998073732497,
|
| 1843 |
+
"loss": 3.0854,
|
| 1844 |
+
"step": 2620
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.048822101187741974,
|
| 1848 |
+
"grad_norm": 6.780309200286865,
|
| 1849 |
+
"learning_rate": 0.00029999998049774177,
|
| 1850 |
+
"loss": 3.3289,
|
| 1851 |
+
"step": 2630
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 0.049347070017717695,
|
| 1855 |
+
"grad_norm": 6.212784290313721,
|
| 1856 |
+
"learning_rate": 0.0002999999802566778,
|
| 1857 |
+
"loss": 3.0708,
|
| 1858 |
+
"step": 2640
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.049872038847693416,
|
| 1862 |
+
"grad_norm": 6.844520092010498,
|
| 1863 |
+
"learning_rate": 0.00029999998001413317,
|
| 1864 |
+
"loss": 3.2822,
|
| 1865 |
+
"step": 2650
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 0.05039700767766914,
|
| 1869 |
+
"grad_norm": 6.3941521644592285,
|
| 1870 |
+
"learning_rate": 0.0002999999797701077,
|
| 1871 |
+
"loss": 3.2076,
|
| 1872 |
+
"step": 2660
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 0.05092197650764486,
|
| 1876 |
+
"grad_norm": 6.843677997589111,
|
| 1877 |
+
"learning_rate": 0.0002999999795246016,
|
| 1878 |
+
"loss": 3.2694,
|
| 1879 |
+
"step": 2670
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.05144694533762058,
|
| 1883 |
+
"grad_norm": 6.763852119445801,
|
| 1884 |
+
"learning_rate": 0.0002999999792776147,
|
| 1885 |
+
"loss": 3.2928,
|
| 1886 |
+
"step": 2680
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.0519719141675963,
|
| 1890 |
+
"grad_norm": 7.747461795806885,
|
| 1891 |
+
"learning_rate": 0.00029999997902914704,
|
| 1892 |
+
"loss": 3.3835,
|
| 1893 |
+
"step": 2690
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.05249688299757202,
|
| 1897 |
+
"grad_norm": 6.702919006347656,
|
| 1898 |
+
"learning_rate": 0.0002999999787791987,
|
| 1899 |
+
"loss": 3.1438,
|
| 1900 |
+
"step": 2700
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 0.05302185182754774,
|
| 1904 |
+
"grad_norm": 7.283872604370117,
|
| 1905 |
+
"learning_rate": 0.0002999999785277696,
|
| 1906 |
+
"loss": 3.3725,
|
| 1907 |
+
"step": 2710
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 0.05354682065752346,
|
| 1911 |
+
"grad_norm": 7.523728847503662,
|
| 1912 |
+
"learning_rate": 0.0002999999782748598,
|
| 1913 |
+
"loss": 3.2487,
|
| 1914 |
+
"step": 2720
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.05407178948749918,
|
| 1918 |
+
"grad_norm": 7.731942653656006,
|
| 1919 |
+
"learning_rate": 0.00029999997802046926,
|
| 1920 |
+
"loss": 3.214,
|
| 1921 |
+
"step": 2730
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 0.0545967583174749,
|
| 1925 |
+
"grad_norm": 7.324973106384277,
|
| 1926 |
+
"learning_rate": 0.0002999999777645979,
|
| 1927 |
+
"loss": 3.1839,
|
| 1928 |
+
"step": 2740
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.05512172714745062,
|
| 1932 |
+
"grad_norm": 7.510282516479492,
|
| 1933 |
+
"learning_rate": 0.0002999999775072459,
|
| 1934 |
+
"loss": 3.2158,
|
| 1935 |
+
"step": 2750
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 0.05564669597742634,
|
| 1939 |
+
"grad_norm": 6.5173211097717285,
|
| 1940 |
+
"learning_rate": 0.0002999999772484131,
|
| 1941 |
+
"loss": 3.0809,
|
| 1942 |
+
"step": 2760
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 0.05617166480740206,
|
| 1946 |
+
"grad_norm": 6.481199741363525,
|
| 1947 |
+
"learning_rate": 0.0002999999769880996,
|
| 1948 |
+
"loss": 3.0955,
|
| 1949 |
+
"step": 2770
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 0.05669663363737778,
|
| 1953 |
+
"grad_norm": 6.510835647583008,
|
| 1954 |
+
"learning_rate": 0.0002999999767263053,
|
| 1955 |
+
"loss": 3.1873,
|
| 1956 |
+
"step": 2780
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 0.0572216024673535,
|
| 1960 |
+
"grad_norm": 7.291550159454346,
|
| 1961 |
+
"learning_rate": 0.00029999997646303035,
|
| 1962 |
+
"loss": 3.0941,
|
| 1963 |
+
"step": 2790
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 0.05774657129732922,
|
| 1967 |
+
"grad_norm": 7.231327056884766,
|
| 1968 |
+
"learning_rate": 0.0002999999761982746,
|
| 1969 |
+
"loss": 3.1549,
|
| 1970 |
+
"step": 2800
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.0005249688299757202,
|
| 1974 |
+
"grad_norm": 6.586388111114502,
|
| 1975 |
+
"learning_rate": 0.00029999997593203816,
|
| 1976 |
+
"loss": 3.161,
|
| 1977 |
+
"step": 2810
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 0.0010499376599514404,
|
| 1981 |
+
"grad_norm": 5.895412445068359,
|
| 1982 |
+
"learning_rate": 0.000299999975664321,
|
| 1983 |
+
"loss": 3.1409,
|
| 1984 |
+
"step": 2820
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 0.0015749064899271605,
|
| 1988 |
+
"grad_norm": 7.71649694442749,
|
| 1989 |
+
"learning_rate": 0.000299999975395123,
|
| 1990 |
+
"loss": 3.1125,
|
| 1991 |
+
"step": 2830
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 0.002099875319902881,
|
| 1995 |
+
"grad_norm": 6.9331512451171875,
|
| 1996 |
+
"learning_rate": 0.00029999997512444437,
|
| 1997 |
+
"loss": 3.2067,
|
| 1998 |
+
"step": 2840
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 0.002624844149878601,
|
| 2002 |
+
"grad_norm": 6.3300676345825195,
|
| 2003 |
+
"learning_rate": 0.00029999997485228495,
|
| 2004 |
+
"loss": 3.1829,
|
| 2005 |
+
"step": 2850
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 0.003149812979854321,
|
| 2009 |
+
"grad_norm": 6.470039367675781,
|
| 2010 |
+
"learning_rate": 0.0002999999745786448,
|
| 2011 |
+
"loss": 3.2371,
|
| 2012 |
+
"step": 2860
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.0036747818098300414,
|
| 2016 |
+
"grad_norm": 6.64292049407959,
|
| 2017 |
+
"learning_rate": 0.00029999997430352394,
|
| 2018 |
+
"loss": 3.153,
|
| 2019 |
+
"step": 2870
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 0.004199750639805762,
|
| 2023 |
+
"grad_norm": 7.108335018157959,
|
| 2024 |
+
"learning_rate": 0.0002999999740269223,
|
| 2025 |
+
"loss": 3.2237,
|
| 2026 |
+
"step": 2880
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.0047247194697814816,
|
| 2030 |
+
"grad_norm": 6.253315448760986,
|
| 2031 |
+
"learning_rate": 0.00029999997374883993,
|
| 2032 |
+
"loss": 3.0427,
|
| 2033 |
+
"step": 2890
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 0.005249688299757202,
|
| 2037 |
+
"grad_norm": 8.03952407836914,
|
| 2038 |
+
"learning_rate": 0.0002999999734692769,
|
| 2039 |
+
"loss": 3.1039,
|
| 2040 |
+
"step": 2900
|
| 2041 |
+
}
|
| 2042 |
+
],
|
| 2043 |
+
"logging_steps": 10,
|
| 2044 |
+
"max_steps": 10000000,
|
| 2045 |
+
"num_input_tokens_seen": 0,
|
| 2046 |
+
"num_train_epochs": 525,
|
| 2047 |
+
"save_steps": 100,
|
| 2048 |
+
"stateful_callbacks": {
|
| 2049 |
+
"TrainerControl": {
|
| 2050 |
+
"args": {
|
| 2051 |
+
"should_epoch_stop": false,
|
| 2052 |
+
"should_evaluate": false,
|
| 2053 |
+
"should_log": false,
|
| 2054 |
+
"should_save": true,
|
| 2055 |
+
"should_training_stop": false
|
| 2056 |
+
},
|
| 2057 |
+
"attributes": {}
|
| 2058 |
+
}
|
| 2059 |
+
},
|
| 2060 |
+
"total_flos": 4.3637936160768e+16,
|
| 2061 |
+
"train_batch_size": 8,
|
| 2062 |
+
"trial_name": null,
|
| 2063 |
+
"trial_params": null
|
| 2064 |
+
}
|
checkpoint-2900/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a40bbf14bc866e7549dd7e35b2f1efa93574f3325f7dcb8eac5341e7f4b4f035
|
| 3 |
+
size 5201
|