Upload folder using huggingface_hub
Browse files- checkpoint-1800/config.json +17 -0
- checkpoint-1800/generation_config.json +6 -0
- checkpoint-1800/hybrid_tokenizer_mapping.json +0 -0
- checkpoint-1800/model.safetensors +3 -0
- checkpoint-1800/optimizer.pt +3 -0
- checkpoint-1800/rng_state.pth +3 -0
- checkpoint-1800/scaler.pt +3 -0
- checkpoint-1800/scheduler.pt +3 -0
- checkpoint-1800/tokenizer_config.json +9 -0
- checkpoint-1800/trainer_state.json +1294 -0
- checkpoint-1800/training_args.bin +3 -0
checkpoint-1800/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_linguistic",
|
| 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-1800/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-1800/hybrid_tokenizer_mapping.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1800/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:681d341ce9f825d31cdd2bfe57f58e7b48df89c9219bf9fca51fabab749cac14
|
| 3 |
+
size 508157712
|
checkpoint-1800/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac59c93cf4360153c44ea86063212037208536ca1be4369d86cc6b94973ac9a6
|
| 3 |
+
size 259122285
|
checkpoint-1800/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:98ac8d1d2524e58441f687ee4006a9a8ea6eab0e8249c4480b06cc6c9fdf6d57
|
| 3 |
+
size 14645
|
checkpoint-1800/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f09ea1c37abc38d4ddbd43b4b4fee9e166e366f04b9ea4c4fe36e18416b298c1
|
| 3 |
+
size 1383
|
checkpoint-1800/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9931e82af62260fa8a065a6990e4820bdf6e0f85e5c660f3f5b151cef7b7ba58
|
| 3 |
+
size 1465
|
checkpoint-1800/tokenizer_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"tokenizer_class": "HFCompatibleLinguisticTokenizer",
|
| 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-1800/trainer_state.json
ADDED
|
@@ -0,0 +1,1294 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.005518126182301645,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1800,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00013795315455754112,
|
| 14 |
+
"grad_norm": 53.4421272277832,
|
| 15 |
+
"learning_rate": 4.1999999999999995e-07,
|
| 16 |
+
"loss": 8.6569,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.00027590630911508224,
|
| 21 |
+
"grad_norm": 57.65703582763672,
|
| 22 |
+
"learning_rate": 1.0199999999999998e-06,
|
| 23 |
+
"loss": 8.6391,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0004138594636726234,
|
| 28 |
+
"grad_norm": 93.79612731933594,
|
| 29 |
+
"learning_rate": 1.62e-06,
|
| 30 |
+
"loss": 8.6097,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.0005518126182301645,
|
| 35 |
+
"grad_norm": 133.8379669189453,
|
| 36 |
+
"learning_rate": 2.22e-06,
|
| 37 |
+
"loss": 8.5995,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0006897657727877056,
|
| 42 |
+
"grad_norm": 99.24571228027344,
|
| 43 |
+
"learning_rate": 2.8199999999999997e-06,
|
| 44 |
+
"loss": 8.5699,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.0008277189273452468,
|
| 49 |
+
"grad_norm": 155.49118041992188,
|
| 50 |
+
"learning_rate": 3.42e-06,
|
| 51 |
+
"loss": 8.5516,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.0009656720819027878,
|
| 56 |
+
"grad_norm": 118.79692840576172,
|
| 57 |
+
"learning_rate": 4.02e-06,
|
| 58 |
+
"loss": 8.5331,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.001103625236460329,
|
| 63 |
+
"grad_norm": 88.51729583740234,
|
| 64 |
+
"learning_rate": 4.62e-06,
|
| 65 |
+
"loss": 8.5212,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0012415783910178701,
|
| 70 |
+
"grad_norm": 117.01810455322266,
|
| 71 |
+
"learning_rate": 5.219999999999999e-06,
|
| 72 |
+
"loss": 8.5007,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0013795315455754113,
|
| 77 |
+
"grad_norm": 104.06542205810547,
|
| 78 |
+
"learning_rate": 5.819999999999999e-06,
|
| 79 |
+
"loss": 8.4948,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.0015174847001329524,
|
| 84 |
+
"grad_norm": 103.02165985107422,
|
| 85 |
+
"learning_rate": 6.4199999999999995e-06,
|
| 86 |
+
"loss": 8.4871,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.0016554378546904936,
|
| 91 |
+
"grad_norm": 119.96959686279297,
|
| 92 |
+
"learning_rate": 7.02e-06,
|
| 93 |
+
"loss": 8.4803,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.0017933910092480345,
|
| 98 |
+
"grad_norm": 83.34615325927734,
|
| 99 |
+
"learning_rate": 7.619999999999999e-06,
|
| 100 |
+
"loss": 8.471,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.0019313441638055756,
|
| 105 |
+
"grad_norm": 76.37083435058594,
|
| 106 |
+
"learning_rate": 8.22e-06,
|
| 107 |
+
"loss": 8.4655,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.002069297318363117,
|
| 112 |
+
"grad_norm": 82.49830627441406,
|
| 113 |
+
"learning_rate": 8.819999999999999e-06,
|
| 114 |
+
"loss": 8.4708,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.002207250472920658,
|
| 119 |
+
"grad_norm": 69.70299530029297,
|
| 120 |
+
"learning_rate": 9.419999999999998e-06,
|
| 121 |
+
"loss": 8.4659,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.0023452036274781993,
|
| 126 |
+
"grad_norm": 103.55642700195312,
|
| 127 |
+
"learning_rate": 1.0019999999999999e-05,
|
| 128 |
+
"loss": 8.4555,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.0024831567820357402,
|
| 133 |
+
"grad_norm": 112.4985122680664,
|
| 134 |
+
"learning_rate": 1.062e-05,
|
| 135 |
+
"loss": 8.4629,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.002621109936593281,
|
| 140 |
+
"grad_norm": 78.48297882080078,
|
| 141 |
+
"learning_rate": 1.122e-05,
|
| 142 |
+
"loss": 8.4562,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.0027590630911508225,
|
| 147 |
+
"grad_norm": 87.78710174560547,
|
| 148 |
+
"learning_rate": 1.1819999999999999e-05,
|
| 149 |
+
"loss": 8.451,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 3.448828863938528e-05,
|
| 154 |
+
"grad_norm": 34.83774948120117,
|
| 155 |
+
"learning_rate": 1.2419999999999998e-05,
|
| 156 |
+
"loss": 8.4432,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 6.897657727877056e-05,
|
| 161 |
+
"grad_norm": 68.05087280273438,
|
| 162 |
+
"learning_rate": 1.3019999999999999e-05,
|
| 163 |
+
"loss": 8.4542,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.00010346486591815585,
|
| 168 |
+
"grad_norm": 46.86287307739258,
|
| 169 |
+
"learning_rate": 1.362e-05,
|
| 170 |
+
"loss": 8.4622,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.00013795315455754112,
|
| 175 |
+
"grad_norm": 38.558746337890625,
|
| 176 |
+
"learning_rate": 1.4219999999999998e-05,
|
| 177 |
+
"loss": 8.4458,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.0001724414431969264,
|
| 182 |
+
"grad_norm": 38.49115753173828,
|
| 183 |
+
"learning_rate": 1.4819999999999999e-05,
|
| 184 |
+
"loss": 8.4495,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.0002069297318363117,
|
| 189 |
+
"grad_norm": 29.765907287597656,
|
| 190 |
+
"learning_rate": 1.5419999999999998e-05,
|
| 191 |
+
"loss": 8.464,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.00024141802047569696,
|
| 196 |
+
"grad_norm": 48.3050537109375,
|
| 197 |
+
"learning_rate": 1.602e-05,
|
| 198 |
+
"loss": 8.4601,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.00027590630911508224,
|
| 203 |
+
"grad_norm": 27.089998245239258,
|
| 204 |
+
"learning_rate": 1.6619999999999997e-05,
|
| 205 |
+
"loss": 8.4628,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.00031039459775446753,
|
| 210 |
+
"grad_norm": 36.98786163330078,
|
| 211 |
+
"learning_rate": 1.7219999999999998e-05,
|
| 212 |
+
"loss": 8.4364,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0003448828863938528,
|
| 217 |
+
"grad_norm": 39.274784088134766,
|
| 218 |
+
"learning_rate": 1.782e-05,
|
| 219 |
+
"loss": 8.4314,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.0003793711750332381,
|
| 224 |
+
"grad_norm": 29.303476333618164,
|
| 225 |
+
"learning_rate": 1.842e-05,
|
| 226 |
+
"loss": 8.4153,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.0004138594636726234,
|
| 231 |
+
"grad_norm": 24.435274124145508,
|
| 232 |
+
"learning_rate": 1.9019999999999997e-05,
|
| 233 |
+
"loss": 8.4489,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.0004483477523120086,
|
| 238 |
+
"grad_norm": 25.8426456451416,
|
| 239 |
+
"learning_rate": 1.962e-05,
|
| 240 |
+
"loss": 8.4297,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.0004828360409513939,
|
| 245 |
+
"grad_norm": 24.164419174194336,
|
| 246 |
+
"learning_rate": 2.022e-05,
|
| 247 |
+
"loss": 8.4398,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.0005173243295907793,
|
| 252 |
+
"grad_norm": 17.73716163635254,
|
| 253 |
+
"learning_rate": 2.082e-05,
|
| 254 |
+
"loss": 8.4333,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.0005518126182301645,
|
| 259 |
+
"grad_norm": 34.02337646484375,
|
| 260 |
+
"learning_rate": 2.1419999999999998e-05,
|
| 261 |
+
"loss": 8.4488,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.0005863009068695498,
|
| 266 |
+
"grad_norm": 26.535123825073242,
|
| 267 |
+
"learning_rate": 2.202e-05,
|
| 268 |
+
"loss": 8.4361,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.0006207891955089351,
|
| 273 |
+
"grad_norm": 32.27824783325195,
|
| 274 |
+
"learning_rate": 2.2619999999999997e-05,
|
| 275 |
+
"loss": 8.4116,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.0006552774841483203,
|
| 280 |
+
"grad_norm": 20.55720329284668,
|
| 281 |
+
"learning_rate": 2.3219999999999998e-05,
|
| 282 |
+
"loss": 8.4151,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.0006897657727877056,
|
| 287 |
+
"grad_norm": 30.87930679321289,
|
| 288 |
+
"learning_rate": 2.382e-05,
|
| 289 |
+
"loss": 8.43,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.0007242540614270909,
|
| 294 |
+
"grad_norm": 19.80055046081543,
|
| 295 |
+
"learning_rate": 2.4419999999999997e-05,
|
| 296 |
+
"loss": 8.4345,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.0007587423500664762,
|
| 301 |
+
"grad_norm": 23.822568893432617,
|
| 302 |
+
"learning_rate": 2.5019999999999998e-05,
|
| 303 |
+
"loss": 8.413,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.0007932306387058614,
|
| 308 |
+
"grad_norm": 24.005403518676758,
|
| 309 |
+
"learning_rate": 2.562e-05,
|
| 310 |
+
"loss": 8.419,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.0008277189273452468,
|
| 315 |
+
"grad_norm": 30.060989379882812,
|
| 316 |
+
"learning_rate": 2.622e-05,
|
| 317 |
+
"loss": 8.4314,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.000862207215984632,
|
| 322 |
+
"grad_norm": 22.84067153930664,
|
| 323 |
+
"learning_rate": 2.6759999999999998e-05,
|
| 324 |
+
"loss": 8.4154,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.0008966955046240172,
|
| 329 |
+
"grad_norm": 22.259994506835938,
|
| 330 |
+
"learning_rate": 2.7299999999999996e-05,
|
| 331 |
+
"loss": 8.4086,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.0009311837932634026,
|
| 336 |
+
"grad_norm": 11.726574897766113,
|
| 337 |
+
"learning_rate": 2.7899999999999997e-05,
|
| 338 |
+
"loss": 8.4284,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.0009656720819027878,
|
| 343 |
+
"grad_norm": 27.165864944458008,
|
| 344 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 345 |
+
"loss": 8.4169,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.001000160370542173,
|
| 350 |
+
"grad_norm": 24.19461441040039,
|
| 351 |
+
"learning_rate": 2.91e-05,
|
| 352 |
+
"loss": 8.4343,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.0010346486591815585,
|
| 357 |
+
"grad_norm": 10.8844633102417,
|
| 358 |
+
"learning_rate": 2.97e-05,
|
| 359 |
+
"loss": 8.4233,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.0010691369478209437,
|
| 364 |
+
"grad_norm": 22.924041748046875,
|
| 365 |
+
"learning_rate": 3.0299999999999998e-05,
|
| 366 |
+
"loss": 8.394,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.001103625236460329,
|
| 371 |
+
"grad_norm": 15.292486190795898,
|
| 372 |
+
"learning_rate": 3.09e-05,
|
| 373 |
+
"loss": 8.4232,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.0011381135250997142,
|
| 378 |
+
"grad_norm": 18.48368263244629,
|
| 379 |
+
"learning_rate": 3.149999999999999e-05,
|
| 380 |
+
"loss": 8.4116,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.0011726018137390997,
|
| 385 |
+
"grad_norm": 17.83504867553711,
|
| 386 |
+
"learning_rate": 3.2099999999999994e-05,
|
| 387 |
+
"loss": 8.4049,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.0012070901023784849,
|
| 392 |
+
"grad_norm": 14.123770713806152,
|
| 393 |
+
"learning_rate": 3.2699999999999995e-05,
|
| 394 |
+
"loss": 8.4031,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.0012415783910178701,
|
| 399 |
+
"grad_norm": 19.69959831237793,
|
| 400 |
+
"learning_rate": 3.3299999999999996e-05,
|
| 401 |
+
"loss": 8.4,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.0012760666796572554,
|
| 406 |
+
"grad_norm": 19.486099243164062,
|
| 407 |
+
"learning_rate": 3.39e-05,
|
| 408 |
+
"loss": 8.4109,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.0013105549682966406,
|
| 413 |
+
"grad_norm": 11.047612190246582,
|
| 414 |
+
"learning_rate": 3.45e-05,
|
| 415 |
+
"loss": 8.4181,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.001345043256936026,
|
| 420 |
+
"grad_norm": 20.63058090209961,
|
| 421 |
+
"learning_rate": 3.51e-05,
|
| 422 |
+
"loss": 8.4011,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.0013795315455754113,
|
| 427 |
+
"grad_norm": 25.09280776977539,
|
| 428 |
+
"learning_rate": 3.5699999999999994e-05,
|
| 429 |
+
"loss": 8.3946,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.0014140198342147965,
|
| 434 |
+
"grad_norm": 15.544500350952148,
|
| 435 |
+
"learning_rate": 3.6299999999999995e-05,
|
| 436 |
+
"loss": 8.4085,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.0014485081228541817,
|
| 441 |
+
"grad_norm": 17.291248321533203,
|
| 442 |
+
"learning_rate": 3.6899999999999996e-05,
|
| 443 |
+
"loss": 8.4186,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.001482996411493567,
|
| 448 |
+
"grad_norm": 20.22525405883789,
|
| 449 |
+
"learning_rate": 3.75e-05,
|
| 450 |
+
"loss": 8.4061,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.0015174847001329524,
|
| 455 |
+
"grad_norm": 11.877947807312012,
|
| 456 |
+
"learning_rate": 3.81e-05,
|
| 457 |
+
"loss": 8.3943,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.0015519729887723376,
|
| 462 |
+
"grad_norm": 15.03212833404541,
|
| 463 |
+
"learning_rate": 3.87e-05,
|
| 464 |
+
"loss": 8.4077,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.0015864612774117229,
|
| 469 |
+
"grad_norm": 11.607198715209961,
|
| 470 |
+
"learning_rate": 3.93e-05,
|
| 471 |
+
"loss": 8.4003,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.0016209495660511081,
|
| 476 |
+
"grad_norm": 16.791051864624023,
|
| 477 |
+
"learning_rate": 3.99e-05,
|
| 478 |
+
"loss": 8.3883,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.0016554378546904936,
|
| 483 |
+
"grad_norm": 13.70166015625,
|
| 484 |
+
"learning_rate": 4.05e-05,
|
| 485 |
+
"loss": 8.4086,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.0016899261433298788,
|
| 490 |
+
"grad_norm": 8.733206748962402,
|
| 491 |
+
"learning_rate": 4.11e-05,
|
| 492 |
+
"loss": 8.3855,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.001724414431969264,
|
| 497 |
+
"grad_norm": 18.953189849853516,
|
| 498 |
+
"learning_rate": 4.17e-05,
|
| 499 |
+
"loss": 8.3839,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.0017589027206086493,
|
| 504 |
+
"grad_norm": 15.253740310668945,
|
| 505 |
+
"learning_rate": 4.229999999999999e-05,
|
| 506 |
+
"loss": 8.4009,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.0017933910092480345,
|
| 511 |
+
"grad_norm": 13.945429801940918,
|
| 512 |
+
"learning_rate": 4.289999999999999e-05,
|
| 513 |
+
"loss": 8.3869,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.00182787929788742,
|
| 518 |
+
"grad_norm": 15.738173484802246,
|
| 519 |
+
"learning_rate": 4.3499999999999993e-05,
|
| 520 |
+
"loss": 8.3847,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.0018623675865268052,
|
| 525 |
+
"grad_norm": 13.38271713256836,
|
| 526 |
+
"learning_rate": 4.4099999999999995e-05,
|
| 527 |
+
"loss": 8.3995,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.0018968558751661904,
|
| 532 |
+
"grad_norm": 14.793967247009277,
|
| 533 |
+
"learning_rate": 4.4699999999999996e-05,
|
| 534 |
+
"loss": 8.3833,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.0019313441638055756,
|
| 539 |
+
"grad_norm": 16.215335845947266,
|
| 540 |
+
"learning_rate": 4.5299999999999997e-05,
|
| 541 |
+
"loss": 8.3903,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.001965832452444961,
|
| 546 |
+
"grad_norm": 21.061214447021484,
|
| 547 |
+
"learning_rate": 4.59e-05,
|
| 548 |
+
"loss": 8.3884,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.002000320741084346,
|
| 553 |
+
"grad_norm": 14.154373168945312,
|
| 554 |
+
"learning_rate": 4.65e-05,
|
| 555 |
+
"loss": 8.4027,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.0020348090297237316,
|
| 560 |
+
"grad_norm": 9.922065734863281,
|
| 561 |
+
"learning_rate": 4.709999999999999e-05,
|
| 562 |
+
"loss": 8.4037,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.002069297318363117,
|
| 567 |
+
"grad_norm": 17.40951919555664,
|
| 568 |
+
"learning_rate": 4.7699999999999994e-05,
|
| 569 |
+
"loss": 8.4032,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.002103785607002502,
|
| 574 |
+
"grad_norm": 17.422048568725586,
|
| 575 |
+
"learning_rate": 4.8299999999999995e-05,
|
| 576 |
+
"loss": 8.3983,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.0021382738956418875,
|
| 581 |
+
"grad_norm": 16.443422317504883,
|
| 582 |
+
"learning_rate": 4.8899999999999996e-05,
|
| 583 |
+
"loss": 8.3957,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.0021727621842812725,
|
| 588 |
+
"grad_norm": 24.262035369873047,
|
| 589 |
+
"learning_rate": 4.95e-05,
|
| 590 |
+
"loss": 8.3895,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.002207250472920658,
|
| 595 |
+
"grad_norm": 13.685011863708496,
|
| 596 |
+
"learning_rate": 5.01e-05,
|
| 597 |
+
"loss": 8.3915,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.0022417387615600434,
|
| 602 |
+
"grad_norm": 10.055377960205078,
|
| 603 |
+
"learning_rate": 5.07e-05,
|
| 604 |
+
"loss": 8.3822,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.0022762270501994284,
|
| 609 |
+
"grad_norm": 13.62060546875,
|
| 610 |
+
"learning_rate": 5.13e-05,
|
| 611 |
+
"loss": 8.3817,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.002310715338838814,
|
| 616 |
+
"grad_norm": 14.18138313293457,
|
| 617 |
+
"learning_rate": 5.1899999999999994e-05,
|
| 618 |
+
"loss": 8.3889,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.0023452036274781993,
|
| 623 |
+
"grad_norm": 6.779985427856445,
|
| 624 |
+
"learning_rate": 5.2499999999999995e-05,
|
| 625 |
+
"loss": 8.3846,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.0023796919161175843,
|
| 630 |
+
"grad_norm": 7.078898906707764,
|
| 631 |
+
"learning_rate": 5.309999999999999e-05,
|
| 632 |
+
"loss": 8.4004,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.0024141802047569698,
|
| 637 |
+
"grad_norm": 14.189363479614258,
|
| 638 |
+
"learning_rate": 5.369999999999999e-05,
|
| 639 |
+
"loss": 8.3998,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.002448668493396355,
|
| 644 |
+
"grad_norm": 16.281265258789062,
|
| 645 |
+
"learning_rate": 5.429999999999999e-05,
|
| 646 |
+
"loss": 8.3943,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.0024831567820357402,
|
| 651 |
+
"grad_norm": 14.237013816833496,
|
| 652 |
+
"learning_rate": 5.489999999999999e-05,
|
| 653 |
+
"loss": 8.382,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.0025176450706751257,
|
| 658 |
+
"grad_norm": 7.917139530181885,
|
| 659 |
+
"learning_rate": 5.5499999999999994e-05,
|
| 660 |
+
"loss": 8.3985,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.0025521333593145107,
|
| 665 |
+
"grad_norm": 13.905786514282227,
|
| 666 |
+
"learning_rate": 5.6099999999999995e-05,
|
| 667 |
+
"loss": 8.3868,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.002586621647953896,
|
| 672 |
+
"grad_norm": 14.011921882629395,
|
| 673 |
+
"learning_rate": 5.6699999999999996e-05,
|
| 674 |
+
"loss": 8.3773,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.002621109936593281,
|
| 679 |
+
"grad_norm": 10.070409774780273,
|
| 680 |
+
"learning_rate": 5.73e-05,
|
| 681 |
+
"loss": 8.3894,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.0026555982252326666,
|
| 686 |
+
"grad_norm": 12.96844482421875,
|
| 687 |
+
"learning_rate": 5.79e-05,
|
| 688 |
+
"loss": 8.3818,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.002690086513872052,
|
| 693 |
+
"grad_norm": 14.524611473083496,
|
| 694 |
+
"learning_rate": 5.85e-05,
|
| 695 |
+
"loss": 8.3793,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.002724574802511437,
|
| 700 |
+
"grad_norm": 23.980628967285156,
|
| 701 |
+
"learning_rate": 5.91e-05,
|
| 702 |
+
"loss": 8.3702,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.0027590630911508225,
|
| 707 |
+
"grad_norm": 79.6168441772461,
|
| 708 |
+
"learning_rate": 5.97e-05,
|
| 709 |
+
"loss": 8.3695,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.0027935513797902075,
|
| 714 |
+
"grad_norm": 11.407526016235352,
|
| 715 |
+
"learning_rate": 6.0299999999999995e-05,
|
| 716 |
+
"loss": 8.3857,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.002828039668429593,
|
| 721 |
+
"grad_norm": 12.697784423828125,
|
| 722 |
+
"learning_rate": 6.0899999999999996e-05,
|
| 723 |
+
"loss": 8.3714,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.0028625279570689784,
|
| 728 |
+
"grad_norm": 7.642915725708008,
|
| 729 |
+
"learning_rate": 6.149999999999999e-05,
|
| 730 |
+
"loss": 8.3924,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.0028970162457083635,
|
| 735 |
+
"grad_norm": 13.278314590454102,
|
| 736 |
+
"learning_rate": 6.209999999999999e-05,
|
| 737 |
+
"loss": 8.3676,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.002931504534347749,
|
| 742 |
+
"grad_norm": 9.300806045532227,
|
| 743 |
+
"learning_rate": 6.269999999999999e-05,
|
| 744 |
+
"loss": 8.3769,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.002965992822987134,
|
| 749 |
+
"grad_norm": 10.766736030578613,
|
| 750 |
+
"learning_rate": 6.33e-05,
|
| 751 |
+
"loss": 8.3722,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.0030004811116265194,
|
| 756 |
+
"grad_norm": 16.214441299438477,
|
| 757 |
+
"learning_rate": 6.39e-05,
|
| 758 |
+
"loss": 8.3866,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.003034969400265905,
|
| 763 |
+
"grad_norm": 12.498828887939453,
|
| 764 |
+
"learning_rate": 6.45e-05,
|
| 765 |
+
"loss": 8.3778,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.00306945768890529,
|
| 770 |
+
"grad_norm": 12.378737449645996,
|
| 771 |
+
"learning_rate": 6.503999999999999e-05,
|
| 772 |
+
"loss": 8.3748,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.0031039459775446753,
|
| 777 |
+
"grad_norm": 7.89847993850708,
|
| 778 |
+
"learning_rate": 6.563999999999999e-05,
|
| 779 |
+
"loss": 8.3751,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.0031384342661840607,
|
| 784 |
+
"grad_norm": 11.73416805267334,
|
| 785 |
+
"learning_rate": 6.623999999999999e-05,
|
| 786 |
+
"loss": 8.3712,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.0031729225548234458,
|
| 791 |
+
"grad_norm": 8.572582244873047,
|
| 792 |
+
"learning_rate": 6.683999999999999e-05,
|
| 793 |
+
"loss": 8.3753,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.003207410843462831,
|
| 798 |
+
"grad_norm": 13.893655776977539,
|
| 799 |
+
"learning_rate": 6.743999999999999e-05,
|
| 800 |
+
"loss": 8.3667,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.0032418991321022162,
|
| 805 |
+
"grad_norm": 7.900502681732178,
|
| 806 |
+
"learning_rate": 6.803999999999999e-05,
|
| 807 |
+
"loss": 8.3636,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.0032763874207416017,
|
| 812 |
+
"grad_norm": 12.325122833251953,
|
| 813 |
+
"learning_rate": 6.864e-05,
|
| 814 |
+
"loss": 8.3722,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.003310875709380987,
|
| 819 |
+
"grad_norm": 12.237879753112793,
|
| 820 |
+
"learning_rate": 6.924e-05,
|
| 821 |
+
"loss": 8.3573,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.003345363998020372,
|
| 826 |
+
"grad_norm": 16.15164566040039,
|
| 827 |
+
"learning_rate": 6.984e-05,
|
| 828 |
+
"loss": 8.3626,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.0033798522866597576,
|
| 833 |
+
"grad_norm": 16.162899017333984,
|
| 834 |
+
"learning_rate": 7.044e-05,
|
| 835 |
+
"loss": 8.3827,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.0034143405752991426,
|
| 840 |
+
"grad_norm": 22.04014015197754,
|
| 841 |
+
"learning_rate": 7.104e-05,
|
| 842 |
+
"loss": 8.3637,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.003448828863938528,
|
| 847 |
+
"grad_norm": 11.339788436889648,
|
| 848 |
+
"learning_rate": 7.164e-05,
|
| 849 |
+
"loss": 8.3725,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.0034833171525779135,
|
| 854 |
+
"grad_norm": 13.985651016235352,
|
| 855 |
+
"learning_rate": 7.223999999999999e-05,
|
| 856 |
+
"loss": 8.3759,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.0035178054412172985,
|
| 861 |
+
"grad_norm": 12.80493450164795,
|
| 862 |
+
"learning_rate": 7.283999999999999e-05,
|
| 863 |
+
"loss": 8.3635,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.003552293729856684,
|
| 868 |
+
"grad_norm": 10.030963897705078,
|
| 869 |
+
"learning_rate": 7.343999999999999e-05,
|
| 870 |
+
"loss": 8.3563,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.003586782018496069,
|
| 875 |
+
"grad_norm": 10.714138984680176,
|
| 876 |
+
"learning_rate": 7.403999999999999e-05,
|
| 877 |
+
"loss": 8.3529,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.0036212703071354544,
|
| 882 |
+
"grad_norm": 14.480472564697266,
|
| 883 |
+
"learning_rate": 7.463999999999999e-05,
|
| 884 |
+
"loss": 8.3377,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.00365575859577484,
|
| 889 |
+
"grad_norm": 6.988164901733398,
|
| 890 |
+
"learning_rate": 7.524e-05,
|
| 891 |
+
"loss": 8.3632,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.003690246884414225,
|
| 896 |
+
"grad_norm": 9.82447338104248,
|
| 897 |
+
"learning_rate": 7.584e-05,
|
| 898 |
+
"loss": 8.3417,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.0037247351730536104,
|
| 903 |
+
"grad_norm": 11.450285911560059,
|
| 904 |
+
"learning_rate": 7.644e-05,
|
| 905 |
+
"loss": 8.3563,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.003759223461692996,
|
| 910 |
+
"grad_norm": 10.065130233764648,
|
| 911 |
+
"learning_rate": 7.703999999999998e-05,
|
| 912 |
+
"loss": 8.3482,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.003793711750332381,
|
| 917 |
+
"grad_norm": 13.670631408691406,
|
| 918 |
+
"learning_rate": 7.763999999999998e-05,
|
| 919 |
+
"loss": 8.3319,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.0038282000389717663,
|
| 924 |
+
"grad_norm": 11.154393196105957,
|
| 925 |
+
"learning_rate": 7.823999999999998e-05,
|
| 926 |
+
"loss": 8.3378,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.0038626883276111513,
|
| 931 |
+
"grad_norm": 13.057133674621582,
|
| 932 |
+
"learning_rate": 7.883999999999998e-05,
|
| 933 |
+
"loss": 8.3481,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.0038971766162505367,
|
| 938 |
+
"grad_norm": 8.13063907623291,
|
| 939 |
+
"learning_rate": 7.943999999999998e-05,
|
| 940 |
+
"loss": 8.3619,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.003931664904889922,
|
| 945 |
+
"grad_norm": 7.690333843231201,
|
| 946 |
+
"learning_rate": 8.003999999999999e-05,
|
| 947 |
+
"loss": 8.3349,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.003966153193529307,
|
| 952 |
+
"grad_norm": 8.945077896118164,
|
| 953 |
+
"learning_rate": 8.063999999999999e-05,
|
| 954 |
+
"loss": 8.3609,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.004000641482168692,
|
| 959 |
+
"grad_norm": 11.839431762695312,
|
| 960 |
+
"learning_rate": 8.123999999999999e-05,
|
| 961 |
+
"loss": 8.3552,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.004035129770808078,
|
| 966 |
+
"grad_norm": 9.798072814941406,
|
| 967 |
+
"learning_rate": 8.183999999999999e-05,
|
| 968 |
+
"loss": 8.3124,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.004069618059447463,
|
| 973 |
+
"grad_norm": 13.58779239654541,
|
| 974 |
+
"learning_rate": 8.243999999999999e-05,
|
| 975 |
+
"loss": 8.3544,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.004104106348086848,
|
| 980 |
+
"grad_norm": 12.656436920166016,
|
| 981 |
+
"learning_rate": 8.303999999999999e-05,
|
| 982 |
+
"loss": 8.3457,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.004138594636726234,
|
| 987 |
+
"grad_norm": 9.099520683288574,
|
| 988 |
+
"learning_rate": 8.363999999999999e-05,
|
| 989 |
+
"loss": 8.3605,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.004173082925365619,
|
| 994 |
+
"grad_norm": 19.523441314697266,
|
| 995 |
+
"learning_rate": 8.423999999999999e-05,
|
| 996 |
+
"loss": 8.3333,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.004207571214005004,
|
| 1001 |
+
"grad_norm": 13.605344772338867,
|
| 1002 |
+
"learning_rate": 8.484e-05,
|
| 1003 |
+
"loss": 8.3487,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.00424205950264439,
|
| 1008 |
+
"grad_norm": 15.3053560256958,
|
| 1009 |
+
"learning_rate": 8.544e-05,
|
| 1010 |
+
"loss": 8.3397,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.004276547791283775,
|
| 1015 |
+
"grad_norm": 7.251153469085693,
|
| 1016 |
+
"learning_rate": 8.604e-05,
|
| 1017 |
+
"loss": 8.349,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.00431103607992316,
|
| 1022 |
+
"grad_norm": 9.477354049682617,
|
| 1023 |
+
"learning_rate": 8.664e-05,
|
| 1024 |
+
"loss": 8.3419,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.004345524368562545,
|
| 1029 |
+
"grad_norm": 14.80703067779541,
|
| 1030 |
+
"learning_rate": 8.724e-05,
|
| 1031 |
+
"loss": 8.319,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.004380012657201931,
|
| 1036 |
+
"grad_norm": 14.25675106048584,
|
| 1037 |
+
"learning_rate": 8.784e-05,
|
| 1038 |
+
"loss": 8.35,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.004414500945841316,
|
| 1043 |
+
"grad_norm": 10.686800003051758,
|
| 1044 |
+
"learning_rate": 8.844e-05,
|
| 1045 |
+
"loss": 8.3404,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.004448989234480701,
|
| 1050 |
+
"grad_norm": 9.987860679626465,
|
| 1051 |
+
"learning_rate": 8.904e-05,
|
| 1052 |
+
"loss": 8.3332,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.004483477523120087,
|
| 1057 |
+
"grad_norm": 7.363917350769043,
|
| 1058 |
+
"learning_rate": 8.963999999999999e-05,
|
| 1059 |
+
"loss": 8.3299,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.004517965811759472,
|
| 1064 |
+
"grad_norm": 14.461793899536133,
|
| 1065 |
+
"learning_rate": 9.023999999999999e-05,
|
| 1066 |
+
"loss": 8.3306,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.004552454100398857,
|
| 1071 |
+
"grad_norm": 8.630695343017578,
|
| 1072 |
+
"learning_rate": 9.083999999999999e-05,
|
| 1073 |
+
"loss": 8.3417,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.004586942389038243,
|
| 1078 |
+
"grad_norm": 8.256804466247559,
|
| 1079 |
+
"learning_rate": 9.143999999999999e-05,
|
| 1080 |
+
"loss": 8.3341,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.004621430677677628,
|
| 1085 |
+
"grad_norm": 13.018980026245117,
|
| 1086 |
+
"learning_rate": 9.203999999999999e-05,
|
| 1087 |
+
"loss": 8.3408,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.004655918966317013,
|
| 1092 |
+
"grad_norm": 8.717267036437988,
|
| 1093 |
+
"learning_rate": 9.264e-05,
|
| 1094 |
+
"loss": 8.3248,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.004690407254956399,
|
| 1099 |
+
"grad_norm": 15.733688354492188,
|
| 1100 |
+
"learning_rate": 9.324e-05,
|
| 1101 |
+
"loss": 8.3205,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.004724895543595784,
|
| 1106 |
+
"grad_norm": 15.750919342041016,
|
| 1107 |
+
"learning_rate": 9.384e-05,
|
| 1108 |
+
"loss": 8.3227,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.004759383832235169,
|
| 1113 |
+
"grad_norm": 12.715235710144043,
|
| 1114 |
+
"learning_rate": 9.444e-05,
|
| 1115 |
+
"loss": 8.3421,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.004793872120874554,
|
| 1120 |
+
"grad_norm": 12.098243713378906,
|
| 1121 |
+
"learning_rate": 9.504e-05,
|
| 1122 |
+
"loss": 8.3332,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.0048283604095139395,
|
| 1127 |
+
"grad_norm": 8.364949226379395,
|
| 1128 |
+
"learning_rate": 9.563999999999999e-05,
|
| 1129 |
+
"loss": 8.3246,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.0048628486981533246,
|
| 1134 |
+
"grad_norm": 12.680130004882812,
|
| 1135 |
+
"learning_rate": 9.623999999999999e-05,
|
| 1136 |
+
"loss": 8.3347,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.00489733698679271,
|
| 1141 |
+
"grad_norm": 5.828690052032471,
|
| 1142 |
+
"learning_rate": 9.683999999999999e-05,
|
| 1143 |
+
"loss": 8.3288,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.0049318252754320955,
|
| 1148 |
+
"grad_norm": 10.879535675048828,
|
| 1149 |
+
"learning_rate": 9.743999999999999e-05,
|
| 1150 |
+
"loss": 8.3261,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.0049663135640714805,
|
| 1155 |
+
"grad_norm": 10.933409690856934,
|
| 1156 |
+
"learning_rate": 9.803999999999999e-05,
|
| 1157 |
+
"loss": 8.3129,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.0050008018527108655,
|
| 1162 |
+
"grad_norm": 9.908263206481934,
|
| 1163 |
+
"learning_rate": 9.863999999999999e-05,
|
| 1164 |
+
"loss": 8.3215,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.005035290141350251,
|
| 1169 |
+
"grad_norm": 9.367692947387695,
|
| 1170 |
+
"learning_rate": 9.923999999999999e-05,
|
| 1171 |
+
"loss": 8.3161,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.005069778429989636,
|
| 1176 |
+
"grad_norm": 17.53960418701172,
|
| 1177 |
+
"learning_rate": 9.983999999999999e-05,
|
| 1178 |
+
"loss": 8.3095,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.005104266718629021,
|
| 1183 |
+
"grad_norm": 12.228901863098145,
|
| 1184 |
+
"learning_rate": 0.00010044,
|
| 1185 |
+
"loss": 8.3064,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.005138755007268406,
|
| 1190 |
+
"grad_norm": 9.411287307739258,
|
| 1191 |
+
"learning_rate": 0.00010103999999999998,
|
| 1192 |
+
"loss": 8.3047,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.005173243295907792,
|
| 1197 |
+
"grad_norm": 12.89146900177002,
|
| 1198 |
+
"learning_rate": 0.00010163999999999998,
|
| 1199 |
+
"loss": 8.313,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.005207731584547177,
|
| 1204 |
+
"grad_norm": 13.037481307983398,
|
| 1205 |
+
"learning_rate": 0.00010223999999999998,
|
| 1206 |
+
"loss": 8.3005,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.005242219873186562,
|
| 1211 |
+
"grad_norm": 18.167877197265625,
|
| 1212 |
+
"learning_rate": 0.00010283999999999998,
|
| 1213 |
+
"loss": 8.3194,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.005276708161825948,
|
| 1218 |
+
"grad_norm": 11.53596019744873,
|
| 1219 |
+
"learning_rate": 0.00010343999999999999,
|
| 1220 |
+
"loss": 8.3103,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.005311196450465333,
|
| 1225 |
+
"grad_norm": 14.443565368652344,
|
| 1226 |
+
"learning_rate": 0.00010403999999999999,
|
| 1227 |
+
"loss": 8.3078,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.005345684739104718,
|
| 1232 |
+
"grad_norm": 12.966043472290039,
|
| 1233 |
+
"learning_rate": 0.00010463999999999999,
|
| 1234 |
+
"loss": 8.3133,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.005380173027744104,
|
| 1239 |
+
"grad_norm": 9.95252513885498,
|
| 1240 |
+
"learning_rate": 0.00010523999999999999,
|
| 1241 |
+
"loss": 8.3033,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.005414661316383489,
|
| 1246 |
+
"grad_norm": 15.110918045043945,
|
| 1247 |
+
"learning_rate": 0.00010583999999999999,
|
| 1248 |
+
"loss": 8.3018,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.005449149605022874,
|
| 1253 |
+
"grad_norm": 9.710939407348633,
|
| 1254 |
+
"learning_rate": 0.00010643999999999999,
|
| 1255 |
+
"loss": 8.3056,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.00548363789366226,
|
| 1260 |
+
"grad_norm": 14.128268241882324,
|
| 1261 |
+
"learning_rate": 0.00010703999999999999,
|
| 1262 |
+
"loss": 8.3116,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.005518126182301645,
|
| 1267 |
+
"grad_norm": 12.291546821594238,
|
| 1268 |
+
"learning_rate": 0.00010763999999999999,
|
| 1269 |
+
"loss": 8.3087,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
}
|
| 1272 |
+
],
|
| 1273 |
+
"logging_steps": 10,
|
| 1274 |
+
"max_steps": 10000000,
|
| 1275 |
+
"num_input_tokens_seen": 0,
|
| 1276 |
+
"num_train_epochs": 35,
|
| 1277 |
+
"save_steps": 100,
|
| 1278 |
+
"stateful_callbacks": {
|
| 1279 |
+
"TrainerControl": {
|
| 1280 |
+
"args": {
|
| 1281 |
+
"should_epoch_stop": false,
|
| 1282 |
+
"should_evaluate": false,
|
| 1283 |
+
"should_log": false,
|
| 1284 |
+
"should_save": true,
|
| 1285 |
+
"should_training_stop": false
|
| 1286 |
+
},
|
| 1287 |
+
"attributes": {}
|
| 1288 |
+
}
|
| 1289 |
+
},
|
| 1290 |
+
"total_flos": 9389851924856832.0,
|
| 1291 |
+
"train_batch_size": 8,
|
| 1292 |
+
"trial_name": null,
|
| 1293 |
+
"trial_params": null
|
| 1294 |
+
}
|
checkpoint-1800/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a51182451ce226f524b4245bb98c4f2eb82c17aaf6895f5b914588d78fefda14
|
| 3 |
+
size 5201
|