Training in progress, step 34000, checkpoint
Browse files- last-checkpoint/config.json +33 -0
- last-checkpoint/model.safetensors +3 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/rng_state.pth +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/tokenizer.json +0 -0
- last-checkpoint/tokenizer_config.json +14 -0
- last-checkpoint/trainer_state.json +2720 -0
- last-checkpoint/training_args.bin +3 -0
last-checkpoint/config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"HybridFourierLM"
|
| 4 |
+
],
|
| 5 |
+
"bos_token_id": 0,
|
| 6 |
+
"dropout": 0.05,
|
| 7 |
+
"dtype": "float32",
|
| 8 |
+
"eos_token_id": 0,
|
| 9 |
+
"latent_dim": 768,
|
| 10 |
+
"layer_types": [
|
| 11 |
+
"linear",
|
| 12 |
+
"linear",
|
| 13 |
+
"linear",
|
| 14 |
+
"softmax",
|
| 15 |
+
"linear",
|
| 16 |
+
"linear",
|
| 17 |
+
"linear",
|
| 18 |
+
"softmax",
|
| 19 |
+
"linear",
|
| 20 |
+
"linear",
|
| 21 |
+
"linear",
|
| 22 |
+
"softmax"
|
| 23 |
+
],
|
| 24 |
+
"model_type": "hybrid_fourier_lm",
|
| 25 |
+
"num_layers": 12,
|
| 26 |
+
"num_modes": 64,
|
| 27 |
+
"pad_token_id": 1,
|
| 28 |
+
"tie_word_embeddings": true,
|
| 29 |
+
"time_scale": 128.0,
|
| 30 |
+
"transformers_version": "5.13.0",
|
| 31 |
+
"use_cache": false,
|
| 32 |
+
"vocab_size": 50277
|
| 33 |
+
}
|
last-checkpoint/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:730a84ab942781715590be7c04cad100c356ef27ced617f53c7a7afc6f79f76e
|
| 3 |
+
size 579824888
|
last-checkpoint/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b801cdd780ceaa960f46aaa0c82910a2ee0f77af3430455271bf70ac88460e56
|
| 3 |
+
size 1159794763
|
last-checkpoint/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3184be6e7aa9630d2d04cfd8a5eadab72c53048d1d80c9edf3ff32dd527e46ff
|
| 3 |
+
size 14645
|
last-checkpoint/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b6825ef0f25e086c9dfb4332e3b7e10288e902088c249c87e47f6e24a13965ac
|
| 3 |
+
size 1465
|
last-checkpoint/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
last-checkpoint/tokenizer_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|endoftext|>",
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"errors": "replace",
|
| 7 |
+
"is_local": false,
|
| 8 |
+
"local_files_only": false,
|
| 9 |
+
"model_max_length": 1000000000,
|
| 10 |
+
"pad_token": "<|padding|>",
|
| 11 |
+
"tokenizer_class": "GPTNeoXTokenizer",
|
| 12 |
+
"trim_offsets": true,
|
| 13 |
+
"unk_token": "<|endoftext|>"
|
| 14 |
+
}
|
last-checkpoint/trainer_state.json
ADDED
|
@@ -0,0 +1,2720 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.04,
|
| 6 |
+
"eval_steps": 1000,
|
| 7 |
+
"global_step": 34000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.002,
|
| 14 |
+
"grad_norm": 0.5271317362785339,
|
| 15 |
+
"learning_rate": 0.00066,
|
| 16 |
+
"loss": 7.779569091796875,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.004,
|
| 21 |
+
"grad_norm": 0.6387752890586853,
|
| 22 |
+
"learning_rate": 0.0009999976160274871,
|
| 23 |
+
"loss": 6.24430908203125,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.006,
|
| 28 |
+
"grad_norm": 0.42202863097190857,
|
| 29 |
+
"learning_rate": 0.000999977956590203,
|
| 30 |
+
"loss": 5.698394165039063,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.008,
|
| 35 |
+
"grad_norm": 0.42532625794410706,
|
| 36 |
+
"learning_rate": 0.0009999384399149713,
|
| 37 |
+
"loss": 5.405557861328125,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.01,
|
| 42 |
+
"grad_norm": 0.4566122889518738,
|
| 43 |
+
"learning_rate": 0.0009998790675712448,
|
| 44 |
+
"loss": 5.189588623046875,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.012,
|
| 49 |
+
"grad_norm": 0.4048032760620117,
|
| 50 |
+
"learning_rate": 0.000999799841917069,
|
| 51 |
+
"loss": 5.048131713867187,
|
| 52 |
+
"step": 600
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.014,
|
| 56 |
+
"grad_norm": 0.33951064944267273,
|
| 57 |
+
"learning_rate": 0.0009997007660989881,
|
| 58 |
+
"loss": 4.923414611816407,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.016,
|
| 63 |
+
"grad_norm": 0.3192354440689087,
|
| 64 |
+
"learning_rate": 0.0009995818440519195,
|
| 65 |
+
"loss": 4.7589013671875,
|
| 66 |
+
"step": 800
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.018,
|
| 70 |
+
"grad_norm": 0.4026377201080322,
|
| 71 |
+
"learning_rate": 0.0009994430804989986,
|
| 72 |
+
"loss": 4.643058166503907,
|
| 73 |
+
"step": 900
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.02,
|
| 77 |
+
"grad_norm": 0.3453679084777832,
|
| 78 |
+
"learning_rate": 0.0009992844809513896,
|
| 79 |
+
"loss": 4.579913330078125,
|
| 80 |
+
"step": 1000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.02,
|
| 84 |
+
"eval_accuracy": 0.269652397386301,
|
| 85 |
+
"eval_loss": 4.505138874053955,
|
| 86 |
+
"eval_runtime": 6.375,
|
| 87 |
+
"eval_samples_per_second": 304.469,
|
| 88 |
+
"eval_steps_per_second": 19.137,
|
| 89 |
+
"step": 1000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.022,
|
| 93 |
+
"grad_norm": 0.3148439824581146,
|
| 94 |
+
"learning_rate": 0.0009991060517080686,
|
| 95 |
+
"loss": 4.5075772094726565,
|
| 96 |
+
"step": 1100
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.024,
|
| 100 |
+
"grad_norm": 0.3168841600418091,
|
| 101 |
+
"learning_rate": 0.0009989077998555718,
|
| 102 |
+
"loss": 4.395697326660156,
|
| 103 |
+
"step": 1200
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.026,
|
| 107 |
+
"grad_norm": 0.277462899684906,
|
| 108 |
+
"learning_rate": 0.0009986897332677144,
|
| 109 |
+
"loss": 4.3901171875,
|
| 110 |
+
"step": 1300
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.028,
|
| 114 |
+
"grad_norm": 0.27992936968803406,
|
| 115 |
+
"learning_rate": 0.0009984518606052791,
|
| 116 |
+
"loss": 4.348124084472656,
|
| 117 |
+
"step": 1400
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.03,
|
| 121 |
+
"grad_norm": 0.2656252980232239,
|
| 122 |
+
"learning_rate": 0.0009981941913156701,
|
| 123 |
+
"loss": 4.3058538818359375,
|
| 124 |
+
"step": 1500
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.032,
|
| 128 |
+
"grad_norm": 0.36868757009506226,
|
| 129 |
+
"learning_rate": 0.0009979167356325394,
|
| 130 |
+
"loss": 4.28273681640625,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.034,
|
| 135 |
+
"grad_norm": 0.25333118438720703,
|
| 136 |
+
"learning_rate": 0.0009976195045753797,
|
| 137 |
+
"loss": 4.249351501464844,
|
| 138 |
+
"step": 1700
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.036,
|
| 142 |
+
"grad_norm": 0.2491893470287323,
|
| 143 |
+
"learning_rate": 0.0009973025099490876,
|
| 144 |
+
"loss": 4.22731689453125,
|
| 145 |
+
"step": 1800
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.038,
|
| 149 |
+
"grad_norm": 0.28424108028411865,
|
| 150 |
+
"learning_rate": 0.0009969657643434933,
|
| 151 |
+
"loss": 4.169223022460938,
|
| 152 |
+
"step": 1900
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.04,
|
| 156 |
+
"grad_norm": 0.2666165232658386,
|
| 157 |
+
"learning_rate": 0.0009966092811328612,
|
| 158 |
+
"loss": 4.170299682617188,
|
| 159 |
+
"step": 2000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.04,
|
| 163 |
+
"eval_accuracy": 0.3056205014664501,
|
| 164 |
+
"eval_loss": 4.103638648986816,
|
| 165 |
+
"eval_runtime": 6.566,
|
| 166 |
+
"eval_samples_per_second": 295.612,
|
| 167 |
+
"eval_steps_per_second": 18.58,
|
| 168 |
+
"step": 2000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.042,
|
| 172 |
+
"grad_norm": 0.3338668942451477,
|
| 173 |
+
"learning_rate": 0.0009962330744753604,
|
| 174 |
+
"loss": 4.163348388671875,
|
| 175 |
+
"step": 2100
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 0.044,
|
| 179 |
+
"grad_norm": 0.25480538606643677,
|
| 180 |
+
"learning_rate": 0.0009958371593124988,
|
| 181 |
+
"loss": 4.118557434082032,
|
| 182 |
+
"step": 2200
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.046,
|
| 186 |
+
"grad_norm": 0.2351720631122589,
|
| 187 |
+
"learning_rate": 0.0009954215513685335,
|
| 188 |
+
"loss": 4.1006707763671875,
|
| 189 |
+
"step": 2300
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.048,
|
| 193 |
+
"grad_norm": 0.23063364624977112,
|
| 194 |
+
"learning_rate": 0.0009949862671498437,
|
| 195 |
+
"loss": 4.084933166503906,
|
| 196 |
+
"step": 2400
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.05,
|
| 200 |
+
"grad_norm": 0.270161896944046,
|
| 201 |
+
"learning_rate": 0.000994531323944276,
|
| 202 |
+
"loss": 4.043960876464844,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.052,
|
| 207 |
+
"grad_norm": 0.23590487241744995,
|
| 208 |
+
"learning_rate": 0.0009940567398204583,
|
| 209 |
+
"loss": 4.043964233398437,
|
| 210 |
+
"step": 2600
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.054,
|
| 214 |
+
"grad_norm": 0.2604995369911194,
|
| 215 |
+
"learning_rate": 0.000993562533627081,
|
| 216 |
+
"loss": 4.023601379394531,
|
| 217 |
+
"step": 2700
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.056,
|
| 221 |
+
"grad_norm": 0.21301409602165222,
|
| 222 |
+
"learning_rate": 0.0009930487249921495,
|
| 223 |
+
"loss": 4.006760864257813,
|
| 224 |
+
"step": 2800
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.058,
|
| 228 |
+
"grad_norm": 0.21575306355953217,
|
| 229 |
+
"learning_rate": 0.0009925153343222043,
|
| 230 |
+
"loss": 3.99057861328125,
|
| 231 |
+
"step": 2900
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.06,
|
| 235 |
+
"grad_norm": 0.2405770868062973,
|
| 236 |
+
"learning_rate": 0.00099196238280151,
|
| 237 |
+
"loss": 4.004396667480469,
|
| 238 |
+
"step": 3000
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.06,
|
| 242 |
+
"eval_accuracy": 0.3194532243250246,
|
| 243 |
+
"eval_loss": 3.945106267929077,
|
| 244 |
+
"eval_runtime": 6.283,
|
| 245 |
+
"eval_samples_per_second": 308.927,
|
| 246 |
+
"eval_steps_per_second": 19.417,
|
| 247 |
+
"step": 3000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.062,
|
| 251 |
+
"grad_norm": 0.22380003333091736,
|
| 252 |
+
"learning_rate": 0.0009913898923912148,
|
| 253 |
+
"loss": 3.9952774047851562,
|
| 254 |
+
"step": 3100
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.064,
|
| 258 |
+
"grad_norm": 0.21692751348018646,
|
| 259 |
+
"learning_rate": 0.0009907978858284775,
|
| 260 |
+
"loss": 3.9829229736328124,
|
| 261 |
+
"step": 3200
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.066,
|
| 265 |
+
"grad_norm": 0.22457344830036163,
|
| 266 |
+
"learning_rate": 0.0009901863866255652,
|
| 267 |
+
"loss": 3.9637200927734373,
|
| 268 |
+
"step": 3300
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.068,
|
| 272 |
+
"grad_norm": 0.20820939540863037,
|
| 273 |
+
"learning_rate": 0.0009895554190689191,
|
| 274 |
+
"loss": 3.930269470214844,
|
| 275 |
+
"step": 3400
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.07,
|
| 279 |
+
"grad_norm": 0.1961025595664978,
|
| 280 |
+
"learning_rate": 0.0009889050082181896,
|
| 281 |
+
"loss": 3.91483154296875,
|
| 282 |
+
"step": 3500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.072,
|
| 286 |
+
"grad_norm": 0.23347830772399902,
|
| 287 |
+
"learning_rate": 0.0009882351799052416,
|
| 288 |
+
"loss": 3.8973043823242186,
|
| 289 |
+
"step": 3600
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.074,
|
| 293 |
+
"grad_norm": 0.21384640038013458,
|
| 294 |
+
"learning_rate": 0.0009875459607331285,
|
| 295 |
+
"loss": 3.9072760009765624,
|
| 296 |
+
"step": 3700
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.076,
|
| 300 |
+
"grad_norm": 0.21380040049552917,
|
| 301 |
+
"learning_rate": 0.000986837378075035,
|
| 302 |
+
"loss": 3.8974929809570313,
|
| 303 |
+
"step": 3800
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.078,
|
| 307 |
+
"grad_norm": 0.2199064940214157,
|
| 308 |
+
"learning_rate": 0.0009861094600731911,
|
| 309 |
+
"loss": 3.884527587890625,
|
| 310 |
+
"step": 3900
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.08,
|
| 314 |
+
"grad_norm": 0.2245541661977768,
|
| 315 |
+
"learning_rate": 0.000985362235637753,
|
| 316 |
+
"loss": 3.863822021484375,
|
| 317 |
+
"step": 4000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.08,
|
| 321 |
+
"eval_accuracy": 0.3280885939521158,
|
| 322 |
+
"eval_loss": 3.8499362468719482,
|
| 323 |
+
"eval_runtime": 6.3172,
|
| 324 |
+
"eval_samples_per_second": 307.255,
|
| 325 |
+
"eval_steps_per_second": 19.312,
|
| 326 |
+
"step": 4000
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.082,
|
| 330 |
+
"grad_norm": 0.20376908779144287,
|
| 331 |
+
"learning_rate": 0.000984595734445656,
|
| 332 |
+
"loss": 3.8685012817382813,
|
| 333 |
+
"step": 4100
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.084,
|
| 337 |
+
"grad_norm": 0.18422065675258636,
|
| 338 |
+
"learning_rate": 0.0009838099869394353,
|
| 339 |
+
"loss": 3.853851318359375,
|
| 340 |
+
"step": 4200
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 0.086,
|
| 344 |
+
"grad_norm": 0.19115814566612244,
|
| 345 |
+
"learning_rate": 0.0009830050243260172,
|
| 346 |
+
"loss": 3.8678164672851563,
|
| 347 |
+
"step": 4300
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 0.088,
|
| 351 |
+
"grad_norm": 0.20425236225128174,
|
| 352 |
+
"learning_rate": 0.0009821808785754791,
|
| 353 |
+
"loss": 3.8287237548828124,
|
| 354 |
+
"step": 4400
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.09,
|
| 358 |
+
"grad_norm": 0.1944635510444641,
|
| 359 |
+
"learning_rate": 0.0009813375824197811,
|
| 360 |
+
"loss": 3.8451242065429687,
|
| 361 |
+
"step": 4500
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.092,
|
| 365 |
+
"grad_norm": 0.17947296798229218,
|
| 366 |
+
"learning_rate": 0.0009804751693514645,
|
| 367 |
+
"loss": 3.845400695800781,
|
| 368 |
+
"step": 4600
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.094,
|
| 372 |
+
"grad_norm": 0.20076608657836914,
|
| 373 |
+
"learning_rate": 0.0009795936736223222,
|
| 374 |
+
"loss": 3.8257626342773436,
|
| 375 |
+
"step": 4700
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.096,
|
| 379 |
+
"grad_norm": 0.18554596602916718,
|
| 380 |
+
"learning_rate": 0.0009786931302420386,
|
| 381 |
+
"loss": 3.808670959472656,
|
| 382 |
+
"step": 4800
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 0.098,
|
| 386 |
+
"grad_norm": 0.23913995921611786,
|
| 387 |
+
"learning_rate": 0.000977773574976799,
|
| 388 |
+
"loss": 3.810527038574219,
|
| 389 |
+
"step": 4900
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 0.1,
|
| 393 |
+
"grad_norm": 0.18414103984832764,
|
| 394 |
+
"learning_rate": 0.0009768350443478686,
|
| 395 |
+
"loss": 3.7990948486328127,
|
| 396 |
+
"step": 5000
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 0.1,
|
| 400 |
+
"eval_accuracy": 0.33500293894949945,
|
| 401 |
+
"eval_loss": 3.7702300548553467,
|
| 402 |
+
"eval_runtime": 6.4137,
|
| 403 |
+
"eval_samples_per_second": 302.634,
|
| 404 |
+
"eval_steps_per_second": 19.022,
|
| 405 |
+
"step": 5000
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.102,
|
| 409 |
+
"grad_norm": 0.20944379270076752,
|
| 410 |
+
"learning_rate": 0.0009758775756301432,
|
| 411 |
+
"loss": 3.798116455078125,
|
| 412 |
+
"step": 5100
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 0.104,
|
| 416 |
+
"grad_norm": 0.18056762218475342,
|
| 417 |
+
"learning_rate": 0.0009749012068506673,
|
| 418 |
+
"loss": 3.7997445678710937,
|
| 419 |
+
"step": 5200
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 0.106,
|
| 423 |
+
"grad_norm": 0.20922255516052246,
|
| 424 |
+
"learning_rate": 0.000973905976787125,
|
| 425 |
+
"loss": 3.7765731811523438,
|
| 426 |
+
"step": 5300
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.108,
|
| 430 |
+
"grad_norm": 0.19328580796718597,
|
| 431 |
+
"learning_rate": 0.0009728919249662993,
|
| 432 |
+
"loss": 3.729337463378906,
|
| 433 |
+
"step": 5400
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 0.11,
|
| 437 |
+
"grad_norm": 0.2031419277191162,
|
| 438 |
+
"learning_rate": 0.0009718590916625022,
|
| 439 |
+
"loss": 3.744057922363281,
|
| 440 |
+
"step": 5500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.112,
|
| 444 |
+
"grad_norm": 0.19743800163269043,
|
| 445 |
+
"learning_rate": 0.0009708075178959756,
|
| 446 |
+
"loss": 3.7054071044921875,
|
| 447 |
+
"step": 5600
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.114,
|
| 451 |
+
"grad_norm": 0.1741236299276352,
|
| 452 |
+
"learning_rate": 0.0009697372454312618,
|
| 453 |
+
"loss": 3.699843444824219,
|
| 454 |
+
"step": 5700
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 0.116,
|
| 458 |
+
"grad_norm": 0.1799750030040741,
|
| 459 |
+
"learning_rate": 0.0009686483167755448,
|
| 460 |
+
"loss": 3.7227572631835937,
|
| 461 |
+
"step": 5800
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 0.118,
|
| 465 |
+
"grad_norm": 0.2036087065935135,
|
| 466 |
+
"learning_rate": 0.000967540775176962,
|
| 467 |
+
"loss": 3.6742709350585936,
|
| 468 |
+
"step": 5900
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 0.12,
|
| 472 |
+
"grad_norm": 0.21463191509246826,
|
| 473 |
+
"learning_rate": 0.0009664146646228871,
|
| 474 |
+
"loss": 3.6995980834960935,
|
| 475 |
+
"step": 6000
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 0.12,
|
| 479 |
+
"eval_accuracy": 0.34141821705074654,
|
| 480 |
+
"eval_loss": 3.711214303970337,
|
| 481 |
+
"eval_runtime": 6.3349,
|
| 482 |
+
"eval_samples_per_second": 306.4,
|
| 483 |
+
"eval_steps_per_second": 19.259,
|
| 484 |
+
"step": 6000
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.122,
|
| 488 |
+
"grad_norm": 0.18666808307170868,
|
| 489 |
+
"learning_rate": 0.000965270029838182,
|
| 490 |
+
"loss": 3.6821221923828125,
|
| 491 |
+
"step": 6100
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.124,
|
| 495 |
+
"grad_norm": 0.15751489996910095,
|
| 496 |
+
"learning_rate": 0.0009641069162834215,
|
| 497 |
+
"loss": 3.671753234863281,
|
| 498 |
+
"step": 6200
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.126,
|
| 502 |
+
"grad_norm": 0.22766178846359253,
|
| 503 |
+
"learning_rate": 0.0009629253701530875,
|
| 504 |
+
"loss": 3.6878753662109376,
|
| 505 |
+
"step": 6300
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.128,
|
| 509 |
+
"grad_norm": 0.1677054762840271,
|
| 510 |
+
"learning_rate": 0.0009617254383737343,
|
| 511 |
+
"loss": 3.6619857788085937,
|
| 512 |
+
"step": 6400
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.13,
|
| 516 |
+
"grad_norm": 0.17945168912410736,
|
| 517 |
+
"learning_rate": 0.0009605071686021245,
|
| 518 |
+
"loss": 3.6484750366210936,
|
| 519 |
+
"step": 6500
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.132,
|
| 523 |
+
"grad_norm": 0.17799174785614014,
|
| 524 |
+
"learning_rate": 0.0009592706092233365,
|
| 525 |
+
"loss": 3.6635867309570314,
|
| 526 |
+
"step": 6600
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.134,
|
| 530 |
+
"grad_norm": 0.5586189031600952,
|
| 531 |
+
"learning_rate": 0.0009580158093488436,
|
| 532 |
+
"loss": 3.6489688110351564,
|
| 533 |
+
"step": 6700
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.136,
|
| 537 |
+
"grad_norm": 0.2055712342262268,
|
| 538 |
+
"learning_rate": 0.000956742818814562,
|
| 539 |
+
"loss": 3.6523666381835938,
|
| 540 |
+
"step": 6800
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 0.138,
|
| 544 |
+
"grad_norm": 0.1812128871679306,
|
| 545 |
+
"learning_rate": 0.0009554516881788724,
|
| 546 |
+
"loss": 3.6479287719726563,
|
| 547 |
+
"step": 6900
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 0.14,
|
| 551 |
+
"grad_norm": 0.1664743423461914,
|
| 552 |
+
"learning_rate": 0.0009541424687206124,
|
| 553 |
+
"loss": 3.6319317626953125,
|
| 554 |
+
"step": 7000
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.14,
|
| 558 |
+
"eval_accuracy": 0.3448148965923309,
|
| 559 |
+
"eval_loss": 3.6666343212127686,
|
| 560 |
+
"eval_runtime": 6.38,
|
| 561 |
+
"eval_samples_per_second": 304.231,
|
| 562 |
+
"eval_steps_per_second": 19.122,
|
| 563 |
+
"step": 7000
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.142,
|
| 567 |
+
"grad_norm": 0.15217284858226776,
|
| 568 |
+
"learning_rate": 0.0009528152124370387,
|
| 569 |
+
"loss": 3.6308517456054688,
|
| 570 |
+
"step": 7100
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.144,
|
| 574 |
+
"grad_norm": 0.16507816314697266,
|
| 575 |
+
"learning_rate": 0.0009514699720417631,
|
| 576 |
+
"loss": 3.6217156982421876,
|
| 577 |
+
"step": 7200
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.146,
|
| 581 |
+
"grad_norm": 0.2040642648935318,
|
| 582 |
+
"learning_rate": 0.0009501068009626583,
|
| 583 |
+
"loss": 3.632703857421875,
|
| 584 |
+
"step": 7300
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.148,
|
| 588 |
+
"grad_norm": 0.18437133729457855,
|
| 589 |
+
"learning_rate": 0.0009487257533397361,
|
| 590 |
+
"loss": 3.6346307373046876,
|
| 591 |
+
"step": 7400
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.15,
|
| 595 |
+
"grad_norm": 0.18760915100574493,
|
| 596 |
+
"learning_rate": 0.0009473268840229971,
|
| 597 |
+
"loss": 3.5973056030273436,
|
| 598 |
+
"step": 7500
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.152,
|
| 602 |
+
"grad_norm": 0.17388536036014557,
|
| 603 |
+
"learning_rate": 0.0009459102485702528,
|
| 604 |
+
"loss": 3.611794738769531,
|
| 605 |
+
"step": 7600
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.154,
|
| 609 |
+
"grad_norm": 0.18797264993190765,
|
| 610 |
+
"learning_rate": 0.0009444759032449177,
|
| 611 |
+
"loss": 3.609403381347656,
|
| 612 |
+
"step": 7700
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.156,
|
| 616 |
+
"grad_norm": 0.18140241503715515,
|
| 617 |
+
"learning_rate": 0.0009430239050137766,
|
| 618 |
+
"loss": 3.6027340698242187,
|
| 619 |
+
"step": 7800
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.158,
|
| 623 |
+
"grad_norm": 0.17433640360832214,
|
| 624 |
+
"learning_rate": 0.0009415543115447203,
|
| 625 |
+
"loss": 3.6214892578125,
|
| 626 |
+
"step": 7900
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.16,
|
| 630 |
+
"grad_norm": 0.16605013608932495,
|
| 631 |
+
"learning_rate": 0.0009400671812044565,
|
| 632 |
+
"loss": 3.6061587524414063,
|
| 633 |
+
"step": 8000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.16,
|
| 637 |
+
"eval_accuracy": 0.34762277801806923,
|
| 638 |
+
"eval_loss": 3.640489339828491,
|
| 639 |
+
"eval_runtime": 8.0453,
|
| 640 |
+
"eval_samples_per_second": 241.259,
|
| 641 |
+
"eval_steps_per_second": 15.164,
|
| 642 |
+
"step": 8000
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 0.162,
|
| 646 |
+
"grad_norm": 0.16548939049243927,
|
| 647 |
+
"learning_rate": 0.0009385625730561914,
|
| 648 |
+
"loss": 3.588617248535156,
|
| 649 |
+
"step": 8100
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.164,
|
| 653 |
+
"grad_norm": 0.16542939841747284,
|
| 654 |
+
"learning_rate": 0.0009370405468572831,
|
| 655 |
+
"loss": 3.590567626953125,
|
| 656 |
+
"step": 8200
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.166,
|
| 660 |
+
"grad_norm": 0.18206581473350525,
|
| 661 |
+
"learning_rate": 0.0009355011630568697,
|
| 662 |
+
"loss": 3.636882629394531,
|
| 663 |
+
"step": 8300
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 0.168,
|
| 667 |
+
"grad_norm": 0.2654706835746765,
|
| 668 |
+
"learning_rate": 0.0009339444827934676,
|
| 669 |
+
"loss": 3.6797540283203123,
|
| 670 |
+
"step": 8400
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 0.17,
|
| 674 |
+
"grad_norm": 0.16825009882450104,
|
| 675 |
+
"learning_rate": 0.0009323705678925427,
|
| 676 |
+
"loss": 3.686376953125,
|
| 677 |
+
"step": 8500
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 0.172,
|
| 681 |
+
"grad_norm": 0.18937945365905762,
|
| 682 |
+
"learning_rate": 0.000930779480864057,
|
| 683 |
+
"loss": 3.6942425537109376,
|
| 684 |
+
"step": 8600
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 0.174,
|
| 688 |
+
"grad_norm": 0.19819994270801544,
|
| 689 |
+
"learning_rate": 0.0009291712848999833,
|
| 690 |
+
"loss": 3.6582196044921873,
|
| 691 |
+
"step": 8700
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.176,
|
| 695 |
+
"grad_norm": 0.1557057648897171,
|
| 696 |
+
"learning_rate": 0.0009275460438717975,
|
| 697 |
+
"loss": 3.6659271240234377,
|
| 698 |
+
"step": 8800
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.178,
|
| 702 |
+
"grad_norm": 0.16359280049800873,
|
| 703 |
+
"learning_rate": 0.0009259038223279413,
|
| 704 |
+
"loss": 3.66092529296875,
|
| 705 |
+
"step": 8900
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.18,
|
| 709 |
+
"grad_norm": 0.17897452414035797,
|
| 710 |
+
"learning_rate": 0.0009242446854912575,
|
| 711 |
+
"loss": 3.6642529296875,
|
| 712 |
+
"step": 9000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 0.18,
|
| 716 |
+
"eval_accuracy": 0.3497692697794326,
|
| 717 |
+
"eval_loss": 3.617546319961548,
|
| 718 |
+
"eval_runtime": 6.279,
|
| 719 |
+
"eval_samples_per_second": 309.124,
|
| 720 |
+
"eval_steps_per_second": 19.43,
|
| 721 |
+
"step": 9000
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 0.182,
|
| 725 |
+
"grad_norm": 0.20515520870685577,
|
| 726 |
+
"learning_rate": 0.0009225686992564015,
|
| 727 |
+
"loss": 3.6445053100585936,
|
| 728 |
+
"step": 9100
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.184,
|
| 732 |
+
"grad_norm": 0.16788582503795624,
|
| 733 |
+
"learning_rate": 0.0009208759301872229,
|
| 734 |
+
"loss": 3.657896728515625,
|
| 735 |
+
"step": 9200
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 0.186,
|
| 739 |
+
"grad_norm": 0.1988508403301239,
|
| 740 |
+
"learning_rate": 0.0009191664455141217,
|
| 741 |
+
"loss": 3.616723937988281,
|
| 742 |
+
"step": 9300
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 0.188,
|
| 746 |
+
"grad_norm": 0.1822570413351059,
|
| 747 |
+
"learning_rate": 0.0009174403131313792,
|
| 748 |
+
"loss": 3.6103390502929686,
|
| 749 |
+
"step": 9400
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 0.19,
|
| 753 |
+
"grad_norm": 0.15679553151130676,
|
| 754 |
+
"learning_rate": 0.0009156976015944607,
|
| 755 |
+
"loss": 3.624302978515625,
|
| 756 |
+
"step": 9500
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 0.192,
|
| 760 |
+
"grad_norm": 0.15683937072753906,
|
| 761 |
+
"learning_rate": 0.0009139383801172929,
|
| 762 |
+
"loss": 3.6558917236328123,
|
| 763 |
+
"step": 9600
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 0.194,
|
| 767 |
+
"grad_norm": 0.1553466022014618,
|
| 768 |
+
"learning_rate": 0.0009121627185695152,
|
| 769 |
+
"loss": 3.6198284912109373,
|
| 770 |
+
"step": 9700
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.196,
|
| 774 |
+
"grad_norm": 0.14881721138954163,
|
| 775 |
+
"learning_rate": 0.0009103706874737043,
|
| 776 |
+
"loss": 3.60408203125,
|
| 777 |
+
"step": 9800
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.198,
|
| 781 |
+
"grad_norm": 0.1628589779138565,
|
| 782 |
+
"learning_rate": 0.0009085623580025737,
|
| 783 |
+
"loss": 3.6047940063476562,
|
| 784 |
+
"step": 9900
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 0.2,
|
| 788 |
+
"grad_norm": 0.15589991211891174,
|
| 789 |
+
"learning_rate": 0.0009067378019761472,
|
| 790 |
+
"loss": 3.599223327636719,
|
| 791 |
+
"step": 10000
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 0.2,
|
| 795 |
+
"eval_accuracy": 0.3534593401629882,
|
| 796 |
+
"eval_loss": 3.579674005508423,
|
| 797 |
+
"eval_runtime": 6.3176,
|
| 798 |
+
"eval_samples_per_second": 307.239,
|
| 799 |
+
"eval_steps_per_second": 19.311,
|
| 800 |
+
"step": 10000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.202,
|
| 804 |
+
"grad_norm": 0.18158847093582153,
|
| 805 |
+
"learning_rate": 0.0009048970918589055,
|
| 806 |
+
"loss": 3.607147521972656,
|
| 807 |
+
"step": 10100
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.204,
|
| 811 |
+
"grad_norm": 0.15705344080924988,
|
| 812 |
+
"learning_rate": 0.0009030403007569094,
|
| 813 |
+
"loss": 3.5972610473632813,
|
| 814 |
+
"step": 10200
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.206,
|
| 818 |
+
"grad_norm": 0.18561947345733643,
|
| 819 |
+
"learning_rate": 0.0009011675024148955,
|
| 820 |
+
"loss": 3.5994192504882814,
|
| 821 |
+
"step": 10300
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.208,
|
| 825 |
+
"grad_norm": 0.17375363409519196,
|
| 826 |
+
"learning_rate": 0.0008992787712133476,
|
| 827 |
+
"loss": 3.592384948730469,
|
| 828 |
+
"step": 10400
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.21,
|
| 832 |
+
"grad_norm": 0.14704184234142303,
|
| 833 |
+
"learning_rate": 0.0008973741821655428,
|
| 834 |
+
"loss": 3.5763116455078126,
|
| 835 |
+
"step": 10500
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.212,
|
| 839 |
+
"grad_norm": 0.1663106381893158,
|
| 840 |
+
"learning_rate": 0.0008954538109145713,
|
| 841 |
+
"loss": 3.59366455078125,
|
| 842 |
+
"step": 10600
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.214,
|
| 846 |
+
"grad_norm": 0.15191592276096344,
|
| 847 |
+
"learning_rate": 0.0008935177337303337,
|
| 848 |
+
"loss": 3.5883197021484374,
|
| 849 |
+
"step": 10700
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.216,
|
| 853 |
+
"grad_norm": 0.15566915273666382,
|
| 854 |
+
"learning_rate": 0.0008915660275065107,
|
| 855 |
+
"loss": 3.5549887084960936,
|
| 856 |
+
"step": 10800
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.218,
|
| 860 |
+
"grad_norm": 0.17490389943122864,
|
| 861 |
+
"learning_rate": 0.0008895987697575093,
|
| 862 |
+
"loss": 3.558960266113281,
|
| 863 |
+
"step": 10900
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.22,
|
| 867 |
+
"grad_norm": 0.17277678847312927,
|
| 868 |
+
"learning_rate": 0.0008876160386153846,
|
| 869 |
+
"loss": 3.5671484375,
|
| 870 |
+
"step": 11000
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.22,
|
| 874 |
+
"eval_accuracy": 0.3558709927196726,
|
| 875 |
+
"eval_loss": 3.554290771484375,
|
| 876 |
+
"eval_runtime": 6.8937,
|
| 877 |
+
"eval_samples_per_second": 281.559,
|
| 878 |
+
"eval_steps_per_second": 17.697,
|
| 879 |
+
"step": 11000
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.222,
|
| 883 |
+
"grad_norm": 0.15808428823947906,
|
| 884 |
+
"learning_rate": 0.0008856179128267363,
|
| 885 |
+
"loss": 3.578575439453125,
|
| 886 |
+
"step": 11100
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 0.224,
|
| 890 |
+
"grad_norm": 0.14474111795425415,
|
| 891 |
+
"learning_rate": 0.0008836044717495819,
|
| 892 |
+
"loss": 3.5772747802734375,
|
| 893 |
+
"step": 11200
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 0.226,
|
| 897 |
+
"grad_norm": 0.1413886845111847,
|
| 898 |
+
"learning_rate": 0.0008815757953502035,
|
| 899 |
+
"loss": 3.556199645996094,
|
| 900 |
+
"step": 11300
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 0.228,
|
| 904 |
+
"grad_norm": 0.1502675712108612,
|
| 905 |
+
"learning_rate": 0.0008795319641999735,
|
| 906 |
+
"loss": 3.570118103027344,
|
| 907 |
+
"step": 11400
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 0.23,
|
| 911 |
+
"grad_norm": 0.1517445147037506,
|
| 912 |
+
"learning_rate": 0.0008774730594721534,
|
| 913 |
+
"loss": 3.5563198852539064,
|
| 914 |
+
"step": 11500
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.232,
|
| 918 |
+
"grad_norm": 0.16406166553497314,
|
| 919 |
+
"learning_rate": 0.0008753991629386706,
|
| 920 |
+
"loss": 3.5616604614257814,
|
| 921 |
+
"step": 11600
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 0.234,
|
| 925 |
+
"grad_norm": 0.14764074981212616,
|
| 926 |
+
"learning_rate": 0.0008733103569668701,
|
| 927 |
+
"loss": 3.5495458984375,
|
| 928 |
+
"step": 11700
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 0.236,
|
| 932 |
+
"grad_norm": 0.17058512568473816,
|
| 933 |
+
"learning_rate": 0.0008712067245162439,
|
| 934 |
+
"loss": 3.5244882202148435,
|
| 935 |
+
"step": 11800
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 0.238,
|
| 939 |
+
"grad_norm": 0.16127121448516846,
|
| 940 |
+
"learning_rate": 0.0008690883491351356,
|
| 941 |
+
"loss": 3.5396826171875,
|
| 942 |
+
"step": 11900
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 0.24,
|
| 946 |
+
"grad_norm": 0.14211590588092804,
|
| 947 |
+
"learning_rate": 0.0008669553149574224,
|
| 948 |
+
"loss": 3.522979431152344,
|
| 949 |
+
"step": 12000
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 0.24,
|
| 953 |
+
"eval_accuracy": 0.3578128166428224,
|
| 954 |
+
"eval_loss": 3.5307533740997314,
|
| 955 |
+
"eval_runtime": 6.6725,
|
| 956 |
+
"eval_samples_per_second": 290.897,
|
| 957 |
+
"eval_steps_per_second": 18.284,
|
| 958 |
+
"step": 12000
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 0.242,
|
| 962 |
+
"grad_norm": 0.1405150443315506,
|
| 963 |
+
"learning_rate": 0.0008648077066991736,
|
| 964 |
+
"loss": 3.527177734375,
|
| 965 |
+
"step": 12100
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 0.244,
|
| 969 |
+
"grad_norm": 0.16479545831680298,
|
| 970 |
+
"learning_rate": 0.0008626456096552858,
|
| 971 |
+
"loss": 3.5393338012695312,
|
| 972 |
+
"step": 12200
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 0.246,
|
| 976 |
+
"grad_norm": 0.1583888679742813,
|
| 977 |
+
"learning_rate": 0.0008604691096960961,
|
| 978 |
+
"loss": 3.5796047973632814,
|
| 979 |
+
"step": 12300
|
| 980 |
+
},
|
| 981 |
+
{
|
| 982 |
+
"epoch": 0.248,
|
| 983 |
+
"grad_norm": 0.17304493486881256,
|
| 984 |
+
"learning_rate": 0.0008582782932639702,
|
| 985 |
+
"loss": 3.591195068359375,
|
| 986 |
+
"step": 12400
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.25,
|
| 990 |
+
"grad_norm": 0.15857121348381042,
|
| 991 |
+
"learning_rate": 0.0008560732473698707,
|
| 992 |
+
"loss": 3.584400634765625,
|
| 993 |
+
"step": 12500
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 0.252,
|
| 997 |
+
"grad_norm": 0.14658208191394806,
|
| 998 |
+
"learning_rate": 0.0008538540595899008,
|
| 999 |
+
"loss": 3.589564208984375,
|
| 1000 |
+
"step": 12600
|
| 1001 |
+
},
|
| 1002 |
+
{
|
| 1003 |
+
"epoch": 0.254,
|
| 1004 |
+
"grad_norm": 0.143401101231575,
|
| 1005 |
+
"learning_rate": 0.0008516208180618252,
|
| 1006 |
+
"loss": 3.5824600219726563,
|
| 1007 |
+
"step": 12700
|
| 1008 |
+
},
|
| 1009 |
+
{
|
| 1010 |
+
"epoch": 0.256,
|
| 1011 |
+
"grad_norm": 0.16244202852249146,
|
| 1012 |
+
"learning_rate": 0.0008493736114815718,
|
| 1013 |
+
"loss": 3.574875793457031,
|
| 1014 |
+
"step": 12800
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 0.258,
|
| 1018 |
+
"grad_norm": 0.1520756632089615,
|
| 1019 |
+
"learning_rate": 0.0008471125290997063,
|
| 1020 |
+
"loss": 3.568282165527344,
|
| 1021 |
+
"step": 12900
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.26,
|
| 1025 |
+
"grad_norm": 0.1582706868648529,
|
| 1026 |
+
"learning_rate": 0.00084483766071789,
|
| 1027 |
+
"loss": 3.5930340576171873,
|
| 1028 |
+
"step": 13000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 0.26,
|
| 1032 |
+
"eval_accuracy": 0.35929186944409996,
|
| 1033 |
+
"eval_loss": 3.5190227031707764,
|
| 1034 |
+
"eval_runtime": 6.3791,
|
| 1035 |
+
"eval_samples_per_second": 304.275,
|
| 1036 |
+
"eval_steps_per_second": 19.125,
|
| 1037 |
+
"step": 13000
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 0.262,
|
| 1041 |
+
"grad_norm": 0.17478469014167786,
|
| 1042 |
+
"learning_rate": 0.0008425490966853113,
|
| 1043 |
+
"loss": 3.5665899658203126,
|
| 1044 |
+
"step": 13100
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 0.264,
|
| 1048 |
+
"grad_norm": 0.15228988230228424,
|
| 1049 |
+
"learning_rate": 0.0008402469278950989,
|
| 1050 |
+
"loss": 3.5444100952148436,
|
| 1051 |
+
"step": 13200
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 0.266,
|
| 1055 |
+
"grad_norm": 0.15244163572788239,
|
| 1056 |
+
"learning_rate": 0.0008379312457807102,
|
| 1057 |
+
"loss": 3.5695867919921875,
|
| 1058 |
+
"step": 13300
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.268,
|
| 1062 |
+
"grad_norm": 0.15935847163200378,
|
| 1063 |
+
"learning_rate": 0.0008356021423123017,
|
| 1064 |
+
"loss": 3.5604925537109375,
|
| 1065 |
+
"step": 13400
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.27,
|
| 1069 |
+
"grad_norm": 0.15347428619861603,
|
| 1070 |
+
"learning_rate": 0.000833259709993075,
|
| 1071 |
+
"loss": 3.5551889038085935,
|
| 1072 |
+
"step": 13500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.272,
|
| 1076 |
+
"grad_norm": 0.21459928154945374,
|
| 1077 |
+
"learning_rate": 0.0008309040418556038,
|
| 1078 |
+
"loss": 3.550736389160156,
|
| 1079 |
+
"step": 13600
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.274,
|
| 1083 |
+
"grad_norm": 0.17246893048286438,
|
| 1084 |
+
"learning_rate": 0.0008285352314581378,
|
| 1085 |
+
"loss": 3.5650180053710936,
|
| 1086 |
+
"step": 13700
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 0.276,
|
| 1090 |
+
"grad_norm": 0.16454122960567474,
|
| 1091 |
+
"learning_rate": 0.0008261533728808883,
|
| 1092 |
+
"loss": 3.549154968261719,
|
| 1093 |
+
"step": 13800
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 0.278,
|
| 1097 |
+
"grad_norm": 0.14885519444942474,
|
| 1098 |
+
"learning_rate": 0.000823758560722291,
|
| 1099 |
+
"loss": 3.5430950927734375,
|
| 1100 |
+
"step": 13900
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 0.28,
|
| 1104 |
+
"grad_norm": 0.1521039605140686,
|
| 1105 |
+
"learning_rate": 0.0008213508900952489,
|
| 1106 |
+
"loss": 3.547527160644531,
|
| 1107 |
+
"step": 14000
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 0.28,
|
| 1111 |
+
"eval_accuracy": 0.36159261824608735,
|
| 1112 |
+
"eval_loss": 3.4961202144622803,
|
| 1113 |
+
"eval_runtime": 6.3742,
|
| 1114 |
+
"eval_samples_per_second": 304.507,
|
| 1115 |
+
"eval_steps_per_second": 19.14,
|
| 1116 |
+
"step": 14000
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.282,
|
| 1120 |
+
"grad_norm": 0.1499979943037033,
|
| 1121 |
+
"learning_rate": 0.0008189304566233547,
|
| 1122 |
+
"loss": 3.5474667358398437,
|
| 1123 |
+
"step": 14100
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.284,
|
| 1127 |
+
"grad_norm": 0.1552744209766388,
|
| 1128 |
+
"learning_rate": 0.0008164973564370934,
|
| 1129 |
+
"loss": 3.5074334716796876,
|
| 1130 |
+
"step": 14200
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.286,
|
| 1134 |
+
"grad_norm": 0.13462179899215698,
|
| 1135 |
+
"learning_rate": 0.0008140516861700238,
|
| 1136 |
+
"loss": 3.5272781372070314,
|
| 1137 |
+
"step": 14300
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.288,
|
| 1141 |
+
"grad_norm": 0.1381780356168747,
|
| 1142 |
+
"learning_rate": 0.000811593542954941,
|
| 1143 |
+
"loss": 3.5094314575195313,
|
| 1144 |
+
"step": 14400
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.29,
|
| 1148 |
+
"grad_norm": 0.13239678740501404,
|
| 1149 |
+
"learning_rate": 0.000809123024420019,
|
| 1150 |
+
"loss": 3.5180905151367186,
|
| 1151 |
+
"step": 14500
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.292,
|
| 1155 |
+
"grad_norm": 0.15930698812007904,
|
| 1156 |
+
"learning_rate": 0.0008066402286849317,
|
| 1157 |
+
"loss": 3.5262945556640624,
|
| 1158 |
+
"step": 14600
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.294,
|
| 1162 |
+
"grad_norm": 0.1524200737476349,
|
| 1163 |
+
"learning_rate": 0.0008041452543569583,
|
| 1164 |
+
"loss": 3.541943359375,
|
| 1165 |
+
"step": 14700
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.296,
|
| 1169 |
+
"grad_norm": 0.14040064811706543,
|
| 1170 |
+
"learning_rate": 0.0008016382005270651,
|
| 1171 |
+
"loss": 3.543092041015625,
|
| 1172 |
+
"step": 14800
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.298,
|
| 1176 |
+
"grad_norm": 0.14799822866916656,
|
| 1177 |
+
"learning_rate": 0.0007991191667659708,
|
| 1178 |
+
"loss": 3.5315762329101563,
|
| 1179 |
+
"step": 14900
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.3,
|
| 1183 |
+
"grad_norm": 0.1449238359928131,
|
| 1184 |
+
"learning_rate": 0.0007965882531201913,
|
| 1185 |
+
"loss": 3.522008361816406,
|
| 1186 |
+
"step": 15000
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.3,
|
| 1190 |
+
"eval_accuracy": 0.36332372503531274,
|
| 1191 |
+
"eval_loss": 3.4778010845184326,
|
| 1192 |
+
"eval_runtime": 6.6567,
|
| 1193 |
+
"eval_samples_per_second": 291.588,
|
| 1194 |
+
"eval_steps_per_second": 18.328,
|
| 1195 |
+
"step": 15000
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 0.302,
|
| 1199 |
+
"grad_norm": 0.13593044877052307,
|
| 1200 |
+
"learning_rate": 0.0007940455601080678,
|
| 1201 |
+
"loss": 3.5114361572265627,
|
| 1202 |
+
"step": 15100
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 0.304,
|
| 1206 |
+
"grad_norm": 0.1423630267381668,
|
| 1207 |
+
"learning_rate": 0.0007914911887157724,
|
| 1208 |
+
"loss": 3.5139959716796874,
|
| 1209 |
+
"step": 15200
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 0.306,
|
| 1213 |
+
"grad_norm": 0.15380671620368958,
|
| 1214 |
+
"learning_rate": 0.0007889252403932986,
|
| 1215 |
+
"loss": 3.515833435058594,
|
| 1216 |
+
"step": 15300
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 0.308,
|
| 1220 |
+
"grad_norm": 0.13759265840053558,
|
| 1221 |
+
"learning_rate": 0.0007863478170504326,
|
| 1222 |
+
"loss": 3.528351135253906,
|
| 1223 |
+
"step": 15400
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 0.31,
|
| 1227 |
+
"grad_norm": 0.13887037336826324,
|
| 1228 |
+
"learning_rate": 0.0007837590210527042,
|
| 1229 |
+
"loss": 3.5276870727539062,
|
| 1230 |
+
"step": 15500
|
| 1231 |
+
},
|
| 1232 |
+
{
|
| 1233 |
+
"epoch": 0.312,
|
| 1234 |
+
"grad_norm": 0.15678516030311584,
|
| 1235 |
+
"learning_rate": 0.0007811589552173225,
|
| 1236 |
+
"loss": 3.515467224121094,
|
| 1237 |
+
"step": 15600
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"epoch": 0.314,
|
| 1241 |
+
"grad_norm": 0.2133788764476776,
|
| 1242 |
+
"learning_rate": 0.0007785477228090924,
|
| 1243 |
+
"loss": 3.5256777954101564,
|
| 1244 |
+
"step": 15700
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 0.316,
|
| 1248 |
+
"grad_norm": 0.17379774153232574,
|
| 1249 |
+
"learning_rate": 0.0007759254275363124,
|
| 1250 |
+
"loss": 3.5354928588867187,
|
| 1251 |
+
"step": 15800
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 0.318,
|
| 1255 |
+
"grad_norm": 0.15530315041542053,
|
| 1256 |
+
"learning_rate": 0.0007732921735466564,
|
| 1257 |
+
"loss": 3.5182159423828123,
|
| 1258 |
+
"step": 15900
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 0.32,
|
| 1262 |
+
"grad_norm": 0.14711181819438934,
|
| 1263 |
+
"learning_rate": 0.0007706480654230373,
|
| 1264 |
+
"loss": 3.5050119018554686,
|
| 1265 |
+
"step": 16000
|
| 1266 |
+
},
|
| 1267 |
+
{
|
| 1268 |
+
"epoch": 0.32,
|
| 1269 |
+
"eval_accuracy": 0.3644458693896563,
|
| 1270 |
+
"eval_loss": 3.4653918743133545,
|
| 1271 |
+
"eval_runtime": 6.2866,
|
| 1272 |
+
"eval_samples_per_second": 308.75,
|
| 1273 |
+
"eval_steps_per_second": 19.406,
|
| 1274 |
+
"step": 16000
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 0.322,
|
| 1278 |
+
"grad_norm": 0.16945767402648926,
|
| 1279 |
+
"learning_rate": 0.0007679932081794532,
|
| 1280 |
+
"loss": 3.5018475341796873,
|
| 1281 |
+
"step": 16100
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 0.324,
|
| 1285 |
+
"grad_norm": 0.14158344268798828,
|
| 1286 |
+
"learning_rate": 0.0007653277072568166,
|
| 1287 |
+
"loss": 3.52821044921875,
|
| 1288 |
+
"step": 16200
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 0.326,
|
| 1292 |
+
"grad_norm": 0.15132898092269897,
|
| 1293 |
+
"learning_rate": 0.0007626516685187669,
|
| 1294 |
+
"loss": 3.5037826538085937,
|
| 1295 |
+
"step": 16300
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 0.328,
|
| 1299 |
+
"grad_norm": 0.1444397270679474,
|
| 1300 |
+
"learning_rate": 0.0007599651982474656,
|
| 1301 |
+
"loss": 3.5029248046875,
|
| 1302 |
+
"step": 16400
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 0.33,
|
| 1306 |
+
"grad_norm": 0.15066049993038177,
|
| 1307 |
+
"learning_rate": 0.0007572684031393755,
|
| 1308 |
+
"loss": 3.5109619140625,
|
| 1309 |
+
"step": 16500
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 0.332,
|
| 1313 |
+
"grad_norm": 0.15420712530612946,
|
| 1314 |
+
"learning_rate": 0.0007545613903010231,
|
| 1315 |
+
"loss": 3.498372802734375,
|
| 1316 |
+
"step": 16600
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 0.334,
|
| 1320 |
+
"grad_norm": 0.15669448673725128,
|
| 1321 |
+
"learning_rate": 0.000751844267244744,
|
| 1322 |
+
"loss": 3.504073181152344,
|
| 1323 |
+
"step": 16700
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 0.336,
|
| 1327 |
+
"grad_norm": 0.14457182586193085,
|
| 1328 |
+
"learning_rate": 0.000749117141884414,
|
| 1329 |
+
"loss": 3.5009725952148436,
|
| 1330 |
+
"step": 16800
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 0.338,
|
| 1334 |
+
"grad_norm": 0.1625896543264389,
|
| 1335 |
+
"learning_rate": 0.0007463801225311631,
|
| 1336 |
+
"loss": 3.5116473388671876,
|
| 1337 |
+
"step": 16900
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 0.34,
|
| 1341 |
+
"grad_norm": 0.1407516449689865,
|
| 1342 |
+
"learning_rate": 0.0007436333178890721,
|
| 1343 |
+
"loss": 3.4775790405273437,
|
| 1344 |
+
"step": 17000
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 0.34,
|
| 1348 |
+
"eval_accuracy": 0.3666105090381519,
|
| 1349 |
+
"eval_loss": 3.4458343982696533,
|
| 1350 |
+
"eval_runtime": 6.6119,
|
| 1351 |
+
"eval_samples_per_second": 293.56,
|
| 1352 |
+
"eval_steps_per_second": 18.451,
|
| 1353 |
+
"step": 17000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.342,
|
| 1357 |
+
"grad_norm": 0.15488570928573608,
|
| 1358 |
+
"learning_rate": 0.0007408768370508576,
|
| 1359 |
+
"loss": 3.5057693481445313,
|
| 1360 |
+
"step": 17100
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.344,
|
| 1364 |
+
"grad_norm": 0.13969258964061737,
|
| 1365 |
+
"learning_rate": 0.0007381107894935382,
|
| 1366 |
+
"loss": 3.4655438232421876,
|
| 1367 |
+
"step": 17200
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.346,
|
| 1371 |
+
"grad_norm": 0.15081284940242767,
|
| 1372 |
+
"learning_rate": 0.0007353352850740858,
|
| 1373 |
+
"loss": 3.490411376953125,
|
| 1374 |
+
"step": 17300
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.348,
|
| 1378 |
+
"grad_norm": 0.1477256566286087,
|
| 1379 |
+
"learning_rate": 0.0007325504340250635,
|
| 1380 |
+
"loss": 3.48463134765625,
|
| 1381 |
+
"step": 17400
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.35,
|
| 1385 |
+
"grad_norm": 0.15759490430355072,
|
| 1386 |
+
"learning_rate": 0.0007297563469502479,
|
| 1387 |
+
"loss": 3.5031024169921876,
|
| 1388 |
+
"step": 17500
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.352,
|
| 1392 |
+
"grad_norm": 0.14924432337284088,
|
| 1393 |
+
"learning_rate": 0.0007269531348202344,
|
| 1394 |
+
"loss": 3.5010330200195314,
|
| 1395 |
+
"step": 17600
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.354,
|
| 1399 |
+
"grad_norm": 0.21179653704166412,
|
| 1400 |
+
"learning_rate": 0.0007241409089680324,
|
| 1401 |
+
"loss": 3.484757385253906,
|
| 1402 |
+
"step": 17700
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.356,
|
| 1406 |
+
"grad_norm": 0.1406019777059555,
|
| 1407 |
+
"learning_rate": 0.0007213197810846416,
|
| 1408 |
+
"loss": 3.4941912841796876,
|
| 1409 |
+
"step": 17800
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.358,
|
| 1413 |
+
"grad_norm": 0.18723711371421814,
|
| 1414 |
+
"learning_rate": 0.0007184898632146171,
|
| 1415 |
+
"loss": 3.4815142822265623,
|
| 1416 |
+
"step": 17900
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.36,
|
| 1420 |
+
"grad_norm": 0.15316542983055115,
|
| 1421 |
+
"learning_rate": 0.0007156512677516189,
|
| 1422 |
+
"loss": 3.4828509521484374,
|
| 1423 |
+
"step": 18000
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.36,
|
| 1427 |
+
"eval_accuracy": 0.36743422147076527,
|
| 1428 |
+
"eval_loss": 3.434631824493408,
|
| 1429 |
+
"eval_runtime": 6.7452,
|
| 1430 |
+
"eval_samples_per_second": 287.759,
|
| 1431 |
+
"eval_steps_per_second": 18.087,
|
| 1432 |
+
"step": 18000
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 0.362,
|
| 1436 |
+
"grad_norm": 0.20023606717586517,
|
| 1437 |
+
"learning_rate": 0.0007128041074339486,
|
| 1438 |
+
"loss": 3.4777850341796874,
|
| 1439 |
+
"step": 18100
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 0.364,
|
| 1443 |
+
"grad_norm": 0.4378589689731598,
|
| 1444 |
+
"learning_rate": 0.0007099484953400711,
|
| 1445 |
+
"loss": 3.483775634765625,
|
| 1446 |
+
"step": 18200
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 0.366,
|
| 1450 |
+
"grad_norm": 0.15016524493694305,
|
| 1451 |
+
"learning_rate": 0.0007070845448841241,
|
| 1452 |
+
"loss": 3.471934509277344,
|
| 1453 |
+
"step": 18300
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 0.368,
|
| 1457 |
+
"grad_norm": 0.17178508639335632,
|
| 1458 |
+
"learning_rate": 0.0007042123698114138,
|
| 1459 |
+
"loss": 3.465678405761719,
|
| 1460 |
+
"step": 18400
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 0.37,
|
| 1464 |
+
"grad_norm": 0.17568770051002502,
|
| 1465 |
+
"learning_rate": 0.0007013320841938966,
|
| 1466 |
+
"loss": 3.484932556152344,
|
| 1467 |
+
"step": 18500
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 0.372,
|
| 1471 |
+
"grad_norm": 0.12633204460144043,
|
| 1472 |
+
"learning_rate": 0.0006984438024256499,
|
| 1473 |
+
"loss": 3.4721084594726563,
|
| 1474 |
+
"step": 18600
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 0.374,
|
| 1478 |
+
"grad_norm": 0.13481932878494263,
|
| 1479 |
+
"learning_rate": 0.0006955476392183275,
|
| 1480 |
+
"loss": 3.441580505371094,
|
| 1481 |
+
"step": 18700
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 0.376,
|
| 1485 |
+
"grad_norm": 0.142787367105484,
|
| 1486 |
+
"learning_rate": 0.0006926437095966044,
|
| 1487 |
+
"loss": 3.472591552734375,
|
| 1488 |
+
"step": 18800
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 0.378,
|
| 1492 |
+
"grad_norm": 0.15913543105125427,
|
| 1493 |
+
"learning_rate": 0.000689732128893608,
|
| 1494 |
+
"loss": 3.4664480590820315,
|
| 1495 |
+
"step": 18900
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 0.38,
|
| 1499 |
+
"grad_norm": 0.5166953802108765,
|
| 1500 |
+
"learning_rate": 0.0006868130127463385,
|
| 1501 |
+
"loss": 3.44828125,
|
| 1502 |
+
"step": 19000
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 0.38,
|
| 1506 |
+
"eval_accuracy": 0.36941738224793846,
|
| 1507 |
+
"eval_loss": 3.4189138412475586,
|
| 1508 |
+
"eval_runtime": 6.3949,
|
| 1509 |
+
"eval_samples_per_second": 303.523,
|
| 1510 |
+
"eval_steps_per_second": 19.078,
|
| 1511 |
+
"step": 19000
|
| 1512 |
+
},
|
| 1513 |
+
{
|
| 1514 |
+
"epoch": 0.382,
|
| 1515 |
+
"grad_norm": 0.138035848736763,
|
| 1516 |
+
"learning_rate": 0.0006838864770910745,
|
| 1517 |
+
"loss": 3.4636468505859375,
|
| 1518 |
+
"step": 19100
|
| 1519 |
+
},
|
| 1520 |
+
{
|
| 1521 |
+
"epoch": 0.384,
|
| 1522 |
+
"grad_norm": 0.2230982631444931,
|
| 1523 |
+
"learning_rate": 0.0006809526381587703,
|
| 1524 |
+
"loss": 3.4643569946289063,
|
| 1525 |
+
"step": 19200
|
| 1526 |
+
},
|
| 1527 |
+
{
|
| 1528 |
+
"epoch": 0.386,
|
| 1529 |
+
"grad_norm": 0.12826211750507355,
|
| 1530 |
+
"learning_rate": 0.0006780116124704383,
|
| 1531 |
+
"loss": 3.45958984375,
|
| 1532 |
+
"step": 19300
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 0.388,
|
| 1536 |
+
"grad_norm": 0.1421106457710266,
|
| 1537 |
+
"learning_rate": 0.0006750635168325223,
|
| 1538 |
+
"loss": 3.4553872680664064,
|
| 1539 |
+
"step": 19400
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 0.39,
|
| 1543 |
+
"grad_norm": 0.14258891344070435,
|
| 1544 |
+
"learning_rate": 0.000672108468332257,
|
| 1545 |
+
"loss": 3.4486114501953127,
|
| 1546 |
+
"step": 19500
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 0.392,
|
| 1550 |
+
"grad_norm": 0.24561843276023865,
|
| 1551 |
+
"learning_rate": 0.0006691465843330194,
|
| 1552 |
+
"loss": 3.4674017333984377,
|
| 1553 |
+
"step": 19600
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 0.394,
|
| 1557 |
+
"grad_norm": 0.13396546244621277,
|
| 1558 |
+
"learning_rate": 0.0006661779824696658,
|
| 1559 |
+
"loss": 3.4596041870117187,
|
| 1560 |
+
"step": 19700
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 0.396,
|
| 1564 |
+
"grad_norm": 0.145228311419487,
|
| 1565 |
+
"learning_rate": 0.000663202780643862,
|
| 1566 |
+
"loss": 3.450461120605469,
|
| 1567 |
+
"step": 19800
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"epoch": 0.398,
|
| 1571 |
+
"grad_norm": 0.1595899611711502,
|
| 1572 |
+
"learning_rate": 0.0006602210970193982,
|
| 1573 |
+
"loss": 3.4370431518554687,
|
| 1574 |
+
"step": 19900
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 0.4,
|
| 1578 |
+
"grad_norm": 0.1425502896308899,
|
| 1579 |
+
"learning_rate": 0.0006572330500174975,
|
| 1580 |
+
"loss": 3.441788330078125,
|
| 1581 |
+
"step": 20000
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 0.4,
|
| 1585 |
+
"eval_accuracy": 0.3703691381064293,
|
| 1586 |
+
"eval_loss": 3.4063475131988525,
|
| 1587 |
+
"eval_runtime": 6.6969,
|
| 1588 |
+
"eval_samples_per_second": 289.838,
|
| 1589 |
+
"eval_steps_per_second": 18.218,
|
| 1590 |
+
"step": 20000
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 0.402,
|
| 1594 |
+
"grad_norm": 0.1547059416770935,
|
| 1595 |
+
"learning_rate": 0.0006542387583121131,
|
| 1596 |
+
"loss": 3.441893310546875,
|
| 1597 |
+
"step": 20100
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 0.404,
|
| 1601 |
+
"grad_norm": 0.15452757477760315,
|
| 1602 |
+
"learning_rate": 0.000651238340825213,
|
| 1603 |
+
"loss": 3.434819641113281,
|
| 1604 |
+
"step": 20200
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 0.406,
|
| 1608 |
+
"grad_norm": 0.13765129446983337,
|
| 1609 |
+
"learning_rate": 0.0006482319167220591,
|
| 1610 |
+
"loss": 3.4564175415039062,
|
| 1611 |
+
"step": 20300
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 0.408,
|
| 1615 |
+
"grad_norm": 0.15758726000785828,
|
| 1616 |
+
"learning_rate": 0.0006452196054064732,
|
| 1617 |
+
"loss": 3.4555868530273437,
|
| 1618 |
+
"step": 20400
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 0.41,
|
| 1622 |
+
"grad_norm": 0.3239542543888092,
|
| 1623 |
+
"learning_rate": 0.0006422015265160946,
|
| 1624 |
+
"loss": 3.4481591796875,
|
| 1625 |
+
"step": 20500
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 0.412,
|
| 1629 |
+
"grad_norm": 0.15911568701267242,
|
| 1630 |
+
"learning_rate": 0.0006391777999176294,
|
| 1631 |
+
"loss": 3.44850341796875,
|
| 1632 |
+
"step": 20600
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 0.414,
|
| 1636 |
+
"grad_norm": 0.12856774032115936,
|
| 1637 |
+
"learning_rate": 0.000636148545702089,
|
| 1638 |
+
"loss": 3.438498840332031,
|
| 1639 |
+
"step": 20700
|
| 1640 |
+
},
|
| 1641 |
+
{
|
| 1642 |
+
"epoch": 0.416,
|
| 1643 |
+
"grad_norm": 0.17129512131214142,
|
| 1644 |
+
"learning_rate": 0.000633113884180021,
|
| 1645 |
+
"loss": 3.438189697265625,
|
| 1646 |
+
"step": 20800
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 0.418,
|
| 1650 |
+
"grad_norm": 0.14792455732822418,
|
| 1651 |
+
"learning_rate": 0.0006300739358767311,
|
| 1652 |
+
"loss": 3.425160217285156,
|
| 1653 |
+
"step": 20900
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 0.42,
|
| 1657 |
+
"grad_norm": 0.13070790469646454,
|
| 1658 |
+
"learning_rate": 0.0006270288215274957,
|
| 1659 |
+
"loss": 3.4182763671875,
|
| 1660 |
+
"step": 21000
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 0.42,
|
| 1664 |
+
"eval_accuracy": 0.3719621193102593,
|
| 1665 |
+
"eval_loss": 3.395620584487915,
|
| 1666 |
+
"eval_runtime": 6.7127,
|
| 1667 |
+
"eval_samples_per_second": 289.155,
|
| 1668 |
+
"eval_steps_per_second": 18.175,
|
| 1669 |
+
"step": 21000
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 0.422,
|
| 1673 |
+
"grad_norm": 0.1333521008491516,
|
| 1674 |
+
"learning_rate": 0.0006239786620727669,
|
| 1675 |
+
"loss": 3.440789794921875,
|
| 1676 |
+
"step": 21100
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.424,
|
| 1680 |
+
"grad_norm": 0.14673130214214325,
|
| 1681 |
+
"learning_rate": 0.0006209235786533696,
|
| 1682 |
+
"loss": 3.41100341796875,
|
| 1683 |
+
"step": 21200
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 0.426,
|
| 1687 |
+
"grad_norm": 0.13843247294425964,
|
| 1688 |
+
"learning_rate": 0.00061786369260569,
|
| 1689 |
+
"loss": 3.4241146850585937,
|
| 1690 |
+
"step": 21300
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.428,
|
| 1694 |
+
"grad_norm": 0.15212669968605042,
|
| 1695 |
+
"learning_rate": 0.0006147991254568566,
|
| 1696 |
+
"loss": 3.4158154296875,
|
| 1697 |
+
"step": 21400
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.43,
|
| 1701 |
+
"grad_norm": 0.1734033226966858,
|
| 1702 |
+
"learning_rate": 0.0006117299989199135,
|
| 1703 |
+
"loss": 3.4223089599609375,
|
| 1704 |
+
"step": 21500
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 0.432,
|
| 1708 |
+
"grad_norm": 0.15912015736103058,
|
| 1709 |
+
"learning_rate": 0.0006086564348889863,
|
| 1710 |
+
"loss": 3.405228271484375,
|
| 1711 |
+
"step": 21600
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 0.434,
|
| 1715 |
+
"grad_norm": 0.1569402515888214,
|
| 1716 |
+
"learning_rate": 0.0006055785554344417,
|
| 1717 |
+
"loss": 3.4033087158203124,
|
| 1718 |
+
"step": 21700
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.436,
|
| 1722 |
+
"grad_norm": 0.15008965134620667,
|
| 1723 |
+
"learning_rate": 0.0006024964827980378,
|
| 1724 |
+
"loss": 3.4061663818359373,
|
| 1725 |
+
"step": 21800
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.438,
|
| 1729 |
+
"grad_norm": 0.12902535498142242,
|
| 1730 |
+
"learning_rate": 0.0005994103393880712,
|
| 1731 |
+
"loss": 3.3769439697265624,
|
| 1732 |
+
"step": 21900
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.44,
|
| 1736 |
+
"grad_norm": 0.14630058407783508,
|
| 1737 |
+
"learning_rate": 0.0005963202477745134,
|
| 1738 |
+
"loss": 3.3858111572265623,
|
| 1739 |
+
"step": 22000
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.44,
|
| 1743 |
+
"eval_accuracy": 0.372494457332805,
|
| 1744 |
+
"eval_loss": 3.3869516849517822,
|
| 1745 |
+
"eval_runtime": 6.5171,
|
| 1746 |
+
"eval_samples_per_second": 297.832,
|
| 1747 |
+
"eval_steps_per_second": 18.72,
|
| 1748 |
+
"step": 22000
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 0.442,
|
| 1752 |
+
"grad_norm": 0.13086584210395813,
|
| 1753 |
+
"learning_rate": 0.0005932263306841439,
|
| 1754 |
+
"loss": 3.4057171630859373,
|
| 1755 |
+
"step": 22100
|
| 1756 |
+
},
|
| 1757 |
+
{
|
| 1758 |
+
"epoch": 0.444,
|
| 1759 |
+
"grad_norm": 0.14566738903522491,
|
| 1760 |
+
"learning_rate": 0.0005901287109956757,
|
| 1761 |
+
"loss": 3.374288330078125,
|
| 1762 |
+
"step": 22200
|
| 1763 |
+
},
|
| 1764 |
+
{
|
| 1765 |
+
"epoch": 0.446,
|
| 1766 |
+
"grad_norm": 0.13566701114177704,
|
| 1767 |
+
"learning_rate": 0.0005870275117348753,
|
| 1768 |
+
"loss": 3.3892340087890624,
|
| 1769 |
+
"step": 22300
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 0.448,
|
| 1773 |
+
"grad_norm": 0.18162570893764496,
|
| 1774 |
+
"learning_rate": 0.0005839228560696758,
|
| 1775 |
+
"loss": 3.417158508300781,
|
| 1776 |
+
"step": 22400
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 0.45,
|
| 1780 |
+
"grad_norm": 0.1380554586648941,
|
| 1781 |
+
"learning_rate": 0.0005808148673052863,
|
| 1782 |
+
"loss": 3.385755615234375,
|
| 1783 |
+
"step": 22500
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"epoch": 0.452,
|
| 1787 |
+
"grad_norm": 0.13876518607139587,
|
| 1788 |
+
"learning_rate": 0.0005777036688792934,
|
| 1789 |
+
"loss": 3.3734304809570315,
|
| 1790 |
+
"step": 22600
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 0.454,
|
| 1794 |
+
"grad_norm": 0.1451270580291748,
|
| 1795 |
+
"learning_rate": 0.0005745893843567593,
|
| 1796 |
+
"loss": 3.3767324829101564,
|
| 1797 |
+
"step": 22700
|
| 1798 |
+
},
|
| 1799 |
+
{
|
| 1800 |
+
"epoch": 0.456,
|
| 1801 |
+
"grad_norm": 0.14231646060943604,
|
| 1802 |
+
"learning_rate": 0.0005714721374253151,
|
| 1803 |
+
"loss": 3.3700411987304686,
|
| 1804 |
+
"step": 22800
|
| 1805 |
+
},
|
| 1806 |
+
{
|
| 1807 |
+
"epoch": 0.458,
|
| 1808 |
+
"grad_norm": 0.14160802960395813,
|
| 1809 |
+
"learning_rate": 0.0005683520518902468,
|
| 1810 |
+
"loss": 3.3719024658203125,
|
| 1811 |
+
"step": 22900
|
| 1812 |
+
},
|
| 1813 |
+
{
|
| 1814 |
+
"epoch": 0.46,
|
| 1815 |
+
"grad_norm": 0.15433672070503235,
|
| 1816 |
+
"learning_rate": 0.0005652292516695795,
|
| 1817 |
+
"loss": 3.3654217529296875,
|
| 1818 |
+
"step": 23000
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 0.46,
|
| 1822 |
+
"eval_accuracy": 0.3727878481747762,
|
| 1823 |
+
"eval_loss": 3.3823931217193604,
|
| 1824 |
+
"eval_runtime": 6.745,
|
| 1825 |
+
"eval_samples_per_second": 287.769,
|
| 1826 |
+
"eval_steps_per_second": 18.087,
|
| 1827 |
+
"step": 23000
|
| 1828 |
+
},
|
| 1829 |
+
{
|
| 1830 |
+
"epoch": 0.462,
|
| 1831 |
+
"grad_norm": 0.13699641823768616,
|
| 1832 |
+
"learning_rate": 0.0005621038607891551,
|
| 1833 |
+
"loss": 3.3590658569335936,
|
| 1834 |
+
"step": 23100
|
| 1835 |
+
},
|
| 1836 |
+
{
|
| 1837 |
+
"epoch": 0.464,
|
| 1838 |
+
"grad_norm": 0.14364606142044067,
|
| 1839 |
+
"learning_rate": 0.0005589760033777068,
|
| 1840 |
+
"loss": 3.3595770263671874,
|
| 1841 |
+
"step": 23200
|
| 1842 |
+
},
|
| 1843 |
+
{
|
| 1844 |
+
"epoch": 0.466,
|
| 1845 |
+
"grad_norm": 0.15491677820682526,
|
| 1846 |
+
"learning_rate": 0.0005558458036619291,
|
| 1847 |
+
"loss": 3.3718606567382814,
|
| 1848 |
+
"step": 23300
|
| 1849 |
+
},
|
| 1850 |
+
{
|
| 1851 |
+
"epoch": 0.468,
|
| 1852 |
+
"grad_norm": 0.14689157903194427,
|
| 1853 |
+
"learning_rate": 0.0005527133859615443,
|
| 1854 |
+
"loss": 3.3682122802734376,
|
| 1855 |
+
"step": 23400
|
| 1856 |
+
},
|
| 1857 |
+
{
|
| 1858 |
+
"epoch": 0.47,
|
| 1859 |
+
"grad_norm": 0.2537364363670349,
|
| 1860 |
+
"learning_rate": 0.0005495788746843641,
|
| 1861 |
+
"loss": 3.3676806640625,
|
| 1862 |
+
"step": 23500
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 0.472,
|
| 1866 |
+
"grad_norm": 0.15198677778244019,
|
| 1867 |
+
"learning_rate": 0.0005464423943213493,
|
| 1868 |
+
"loss": 3.3796755981445314,
|
| 1869 |
+
"step": 23600
|
| 1870 |
+
},
|
| 1871 |
+
{
|
| 1872 |
+
"epoch": 0.474,
|
| 1873 |
+
"grad_norm": 0.14707693457603455,
|
| 1874 |
+
"learning_rate": 0.0005433040694416661,
|
| 1875 |
+
"loss": 3.3943267822265626,
|
| 1876 |
+
"step": 23700
|
| 1877 |
+
},
|
| 1878 |
+
{
|
| 1879 |
+
"epoch": 0.476,
|
| 1880 |
+
"grad_norm": 0.13588809967041016,
|
| 1881 |
+
"learning_rate": 0.0005401640246877371,
|
| 1882 |
+
"loss": 3.3692440795898437,
|
| 1883 |
+
"step": 23800
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 0.478,
|
| 1887 |
+
"grad_norm": 0.1468556970357895,
|
| 1888 |
+
"learning_rate": 0.0005370223847702919,
|
| 1889 |
+
"loss": 3.35807861328125,
|
| 1890 |
+
"step": 23900
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"epoch": 0.48,
|
| 1894 |
+
"grad_norm": 0.15797235071659088,
|
| 1895 |
+
"learning_rate": 0.0005338792744634145,
|
| 1896 |
+
"loss": 3.3942901611328127,
|
| 1897 |
+
"step": 24000
|
| 1898 |
+
},
|
| 1899 |
+
{
|
| 1900 |
+
"epoch": 0.48,
|
| 1901 |
+
"eval_accuracy": 0.3733514408918275,
|
| 1902 |
+
"eval_loss": 3.375581741333008,
|
| 1903 |
+
"eval_runtime": 7.2715,
|
| 1904 |
+
"eval_samples_per_second": 266.931,
|
| 1905 |
+
"eval_steps_per_second": 16.778,
|
| 1906 |
+
"step": 24000
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.482,
|
| 1910 |
+
"grad_norm": 0.15001367032527924,
|
| 1911 |
+
"learning_rate": 0.0005307348185995866,
|
| 1912 |
+
"loss": 3.37049560546875,
|
| 1913 |
+
"step": 24100
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.484,
|
| 1917 |
+
"grad_norm": 0.15163348615169525,
|
| 1918 |
+
"learning_rate": 0.0005275891420647307,
|
| 1919 |
+
"loss": 3.3899249267578124,
|
| 1920 |
+
"step": 24200
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.486,
|
| 1924 |
+
"grad_norm": 0.13754494488239288,
|
| 1925 |
+
"learning_rate": 0.0005244423697932499,
|
| 1926 |
+
"loss": 3.397990417480469,
|
| 1927 |
+
"step": 24300
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.488,
|
| 1931 |
+
"grad_norm": 0.13546991348266602,
|
| 1932 |
+
"learning_rate": 0.0005212946267630653,
|
| 1933 |
+
"loss": 3.4154788208007814,
|
| 1934 |
+
"step": 24400
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.49,
|
| 1938 |
+
"grad_norm": 0.17627890408039093,
|
| 1939 |
+
"learning_rate": 0.0005181460379906534,
|
| 1940 |
+
"loss": 3.4125030517578123,
|
| 1941 |
+
"step": 24500
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.492,
|
| 1945 |
+
"grad_norm": 0.14234718680381775,
|
| 1946 |
+
"learning_rate": 0.0005149967285260802,
|
| 1947 |
+
"loss": 3.4010299682617187,
|
| 1948 |
+
"step": 24600
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.494,
|
| 1952 |
+
"grad_norm": 0.1589398831129074,
|
| 1953 |
+
"learning_rate": 0.0005118468234480345,
|
| 1954 |
+
"loss": 3.4132962036132812,
|
| 1955 |
+
"step": 24700
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.496,
|
| 1959 |
+
"grad_norm": 0.12560983002185822,
|
| 1960 |
+
"learning_rate": 0.0005086964478588614,
|
| 1961 |
+
"loss": 3.3963751220703124,
|
| 1962 |
+
"step": 24800
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.498,
|
| 1966 |
+
"grad_norm": 0.14177259802818298,
|
| 1967 |
+
"learning_rate": 0.0005055457268795923,
|
| 1968 |
+
"loss": 3.3922305297851563,
|
| 1969 |
+
"step": 24900
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.5,
|
| 1973 |
+
"grad_norm": 0.13485221564769745,
|
| 1974 |
+
"learning_rate": 0.0005023947856449762,
|
| 1975 |
+
"loss": 3.3956787109375,
|
| 1976 |
+
"step": 25000
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.5,
|
| 1980 |
+
"eval_accuracy": 0.3757247812423439,
|
| 1981 |
+
"eval_loss": 3.3547720909118652,
|
| 1982 |
+
"eval_runtime": 6.4257,
|
| 1983 |
+
"eval_samples_per_second": 302.069,
|
| 1984 |
+
"eval_steps_per_second": 18.986,
|
| 1985 |
+
"step": 25000
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 0.502,
|
| 1989 |
+
"grad_norm": 0.1296602189540863,
|
| 1990 |
+
"learning_rate": 0.00049924374929851,
|
| 1991 |
+
"loss": 3.4078851318359376,
|
| 1992 |
+
"step": 25100
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 0.504,
|
| 1996 |
+
"grad_norm": 0.17173421382904053,
|
| 1997 |
+
"learning_rate": 0.0004960927429874685,
|
| 1998 |
+
"loss": 3.378636779785156,
|
| 1999 |
+
"step": 25200
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 0.506,
|
| 2003 |
+
"grad_norm": 0.14787626266479492,
|
| 2004 |
+
"learning_rate": 0.0004929418918579327,
|
| 2005 |
+
"loss": 3.3894952392578124,
|
| 2006 |
+
"step": 25300
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 0.508,
|
| 2010 |
+
"grad_norm": 0.14193665981292725,
|
| 2011 |
+
"learning_rate": 0.0004897913210498212,
|
| 2012 |
+
"loss": 3.3784576416015626,
|
| 2013 |
+
"step": 25400
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 0.51,
|
| 2017 |
+
"grad_norm": 0.34911948442459106,
|
| 2018 |
+
"learning_rate": 0.0004866411556919189,
|
| 2019 |
+
"loss": 3.4079013061523438,
|
| 2020 |
+
"step": 25500
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 0.512,
|
| 2024 |
+
"grad_norm": 0.14956624805927277,
|
| 2025 |
+
"learning_rate": 0.00048349152089690765,
|
| 2026 |
+
"loss": 3.395030517578125,
|
| 2027 |
+
"step": 25600
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 0.514,
|
| 2031 |
+
"grad_norm": 0.1397118866443634,
|
| 2032 |
+
"learning_rate": 0.0004803425417563974,
|
| 2033 |
+
"loss": 3.3780328369140626,
|
| 2034 |
+
"step": 25700
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 0.516,
|
| 2038 |
+
"grad_norm": 0.13487640023231506,
|
| 2039 |
+
"learning_rate": 0.00047719434333595825,
|
| 2040 |
+
"loss": 3.3928378295898436,
|
| 2041 |
+
"step": 25800
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 0.518,
|
| 2045 |
+
"grad_norm": 0.14225220680236816,
|
| 2046 |
+
"learning_rate": 0.0004740470506701529,
|
| 2047 |
+
"loss": 3.381436767578125,
|
| 2048 |
+
"step": 25900
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 0.52,
|
| 2052 |
+
"grad_norm": 0.13680697977542877,
|
| 2053 |
+
"learning_rate": 0.0004709007887575706,
|
| 2054 |
+
"loss": 3.378739318847656,
|
| 2055 |
+
"step": 26000
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 0.52,
|
| 2059 |
+
"eval_accuracy": 0.3769195171452164,
|
| 2060 |
+
"eval_loss": 3.345224142074585,
|
| 2061 |
+
"eval_runtime": 6.5169,
|
| 2062 |
+
"eval_samples_per_second": 297.843,
|
| 2063 |
+
"eval_steps_per_second": 18.721,
|
| 2064 |
+
"step": 26000
|
| 2065 |
+
},
|
| 2066 |
+
{
|
| 2067 |
+
"epoch": 0.522,
|
| 2068 |
+
"grad_norm": 0.14482583105564117,
|
| 2069 |
+
"learning_rate": 0.0004677556825558636,
|
| 2070 |
+
"loss": 3.380513000488281,
|
| 2071 |
+
"step": 26100
|
| 2072 |
+
},
|
| 2073 |
+
{
|
| 2074 |
+
"epoch": 0.524,
|
| 2075 |
+
"grad_norm": 0.1740926057100296,
|
| 2076 |
+
"learning_rate": 0.0004646118569767831,
|
| 2077 |
+
"loss": 3.3777206420898436,
|
| 2078 |
+
"step": 26200
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 0.526,
|
| 2082 |
+
"grad_norm": 0.13794536888599396,
|
| 2083 |
+
"learning_rate": 0.00046146943688121875,
|
| 2084 |
+
"loss": 3.3788577270507814,
|
| 2085 |
+
"step": 26300
|
| 2086 |
+
},
|
| 2087 |
+
{
|
| 2088 |
+
"epoch": 0.528,
|
| 2089 |
+
"grad_norm": 0.14884917438030243,
|
| 2090 |
+
"learning_rate": 0.00045832854707424044,
|
| 2091 |
+
"loss": 3.372579345703125,
|
| 2092 |
+
"step": 26400
|
| 2093 |
+
},
|
| 2094 |
+
{
|
| 2095 |
+
"epoch": 0.53,
|
| 2096 |
+
"grad_norm": 0.14392748475074768,
|
| 2097 |
+
"learning_rate": 0.0004551893123001399,
|
| 2098 |
+
"loss": 3.369126281738281,
|
| 2099 |
+
"step": 26500
|
| 2100 |
+
},
|
| 2101 |
+
{
|
| 2102 |
+
"epoch": 0.532,
|
| 2103 |
+
"grad_norm": 0.1415095329284668,
|
| 2104 |
+
"learning_rate": 0.0004520518572374776,
|
| 2105 |
+
"loss": 3.4073892211914063,
|
| 2106 |
+
"step": 26600
|
| 2107 |
+
},
|
| 2108 |
+
{
|
| 2109 |
+
"epoch": 0.534,
|
| 2110 |
+
"grad_norm": 0.17241106927394867,
|
| 2111 |
+
"learning_rate": 0.00044891630649413096,
|
| 2112 |
+
"loss": 3.3727621459960937,
|
| 2113 |
+
"step": 26700
|
| 2114 |
+
},
|
| 2115 |
+
{
|
| 2116 |
+
"epoch": 0.536,
|
| 2117 |
+
"grad_norm": 0.1402324140071869,
|
| 2118 |
+
"learning_rate": 0.0004457827846023443,
|
| 2119 |
+
"loss": 3.3563668823242185,
|
| 2120 |
+
"step": 26800
|
| 2121 |
+
},
|
| 2122 |
+
{
|
| 2123 |
+
"epoch": 0.538,
|
| 2124 |
+
"grad_norm": 0.13993528485298157,
|
| 2125 |
+
"learning_rate": 0.0004426514160137837,
|
| 2126 |
+
"loss": 3.36125244140625,
|
| 2127 |
+
"step": 26900
|
| 2128 |
+
},
|
| 2129 |
+
{
|
| 2130 |
+
"epoch": 0.54,
|
| 2131 |
+
"grad_norm": 0.14961598813533783,
|
| 2132 |
+
"learning_rate": 0.000439522325094595,
|
| 2133 |
+
"loss": 3.360479431152344,
|
| 2134 |
+
"step": 27000
|
| 2135 |
+
},
|
| 2136 |
+
{
|
| 2137 |
+
"epoch": 0.54,
|
| 2138 |
+
"eval_accuracy": 0.37769785985999915,
|
| 2139 |
+
"eval_loss": 3.3393123149871826,
|
| 2140 |
+
"eval_runtime": 6.2827,
|
| 2141 |
+
"eval_samples_per_second": 308.945,
|
| 2142 |
+
"eval_steps_per_second": 19.418,
|
| 2143 |
+
"step": 27000
|
| 2144 |
+
},
|
| 2145 |
+
{
|
| 2146 |
+
"epoch": 0.542,
|
| 2147 |
+
"grad_norm": 0.12927019596099854,
|
| 2148 |
+
"learning_rate": 0.0004363956361204626,
|
| 2149 |
+
"loss": 3.3508065795898436,
|
| 2150 |
+
"step": 27100
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 0.544,
|
| 2154 |
+
"grad_norm": 0.15256398916244507,
|
| 2155 |
+
"learning_rate": 0.0004332714732716751,
|
| 2156 |
+
"loss": 3.357308044433594,
|
| 2157 |
+
"step": 27200
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 0.546,
|
| 2161 |
+
"grad_norm": 0.13779006898403168,
|
| 2162 |
+
"learning_rate": 0.0004301499606281931,
|
| 2163 |
+
"loss": 3.3775006103515626,
|
| 2164 |
+
"step": 27300
|
| 2165 |
+
},
|
| 2166 |
+
{
|
| 2167 |
+
"epoch": 0.548,
|
| 2168 |
+
"grad_norm": 0.1694018542766571,
|
| 2169 |
+
"learning_rate": 0.00042703122216472094,
|
| 2170 |
+
"loss": 3.3572964477539062,
|
| 2171 |
+
"step": 27400
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 0.55,
|
| 2175 |
+
"grad_norm": 0.1522066593170166,
|
| 2176 |
+
"learning_rate": 0.00042391538174578263,
|
| 2177 |
+
"loss": 3.3765057373046874,
|
| 2178 |
+
"step": 27500
|
| 2179 |
+
},
|
| 2180 |
+
{
|
| 2181 |
+
"epoch": 0.552,
|
| 2182 |
+
"grad_norm": 0.14023618400096893,
|
| 2183 |
+
"learning_rate": 0.0004208025631208036,
|
| 2184 |
+
"loss": 3.355110778808594,
|
| 2185 |
+
"step": 27600
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 0.554,
|
| 2189 |
+
"grad_norm": 0.14079532027244568,
|
| 2190 |
+
"learning_rate": 0.0004176928899191941,
|
| 2191 |
+
"loss": 3.3566073608398437,
|
| 2192 |
+
"step": 27700
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 0.556,
|
| 2196 |
+
"grad_norm": 0.14286920428276062,
|
| 2197 |
+
"learning_rate": 0.0004145864856454404,
|
| 2198 |
+
"loss": 3.3700027465820312,
|
| 2199 |
+
"step": 27800
|
| 2200 |
+
},
|
| 2201 |
+
{
|
| 2202 |
+
"epoch": 0.558,
|
| 2203 |
+
"grad_norm": 0.13522571325302124,
|
| 2204 |
+
"learning_rate": 0.00041148347367419983,
|
| 2205 |
+
"loss": 3.3883221435546873,
|
| 2206 |
+
"step": 27900
|
| 2207 |
+
},
|
| 2208 |
+
{
|
| 2209 |
+
"epoch": 0.56,
|
| 2210 |
+
"grad_norm": 0.18117037415504456,
|
| 2211 |
+
"learning_rate": 0.00040838397724539947,
|
| 2212 |
+
"loss": 3.3707275390625,
|
| 2213 |
+
"step": 28000
|
| 2214 |
+
},
|
| 2215 |
+
{
|
| 2216 |
+
"epoch": 0.56,
|
| 2217 |
+
"eval_accuracy": 0.3787030511639349,
|
| 2218 |
+
"eval_loss": 3.330246925354004,
|
| 2219 |
+
"eval_runtime": 6.6261,
|
| 2220 |
+
"eval_samples_per_second": 292.932,
|
| 2221 |
+
"eval_steps_per_second": 18.412,
|
| 2222 |
+
"step": 28000
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.562,
|
| 2226 |
+
"grad_norm": 0.14532077312469482,
|
| 2227 |
+
"learning_rate": 0.0004052881194593424,
|
| 2228 |
+
"loss": 3.372833251953125,
|
| 2229 |
+
"step": 28100
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.564,
|
| 2233 |
+
"grad_norm": 0.13675428926944733,
|
| 2234 |
+
"learning_rate": 0.0004021960232718194,
|
| 2235 |
+
"loss": 3.3700936889648436,
|
| 2236 |
+
"step": 28200
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.566,
|
| 2240 |
+
"grad_norm": 0.15135474503040314,
|
| 2241 |
+
"learning_rate": 0.00039910781148922364,
|
| 2242 |
+
"loss": 3.3734249877929687,
|
| 2243 |
+
"step": 28300
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.568,
|
| 2247 |
+
"grad_norm": 0.1487869918346405,
|
| 2248 |
+
"learning_rate": 0.00039602360676367474,
|
| 2249 |
+
"loss": 3.3548593139648437,
|
| 2250 |
+
"step": 28400
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.57,
|
| 2254 |
+
"grad_norm": 0.12845182418823242,
|
| 2255 |
+
"learning_rate": 0.00039294353158814783,
|
| 2256 |
+
"loss": 3.3592022705078124,
|
| 2257 |
+
"step": 28500
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.572,
|
| 2261 |
+
"grad_norm": 0.15905176103115082,
|
| 2262 |
+
"learning_rate": 0.0003898677082916068,
|
| 2263 |
+
"loss": 3.375569152832031,
|
| 2264 |
+
"step": 28600
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.574,
|
| 2268 |
+
"grad_norm": 0.1350318342447281,
|
| 2269 |
+
"learning_rate": 0.0003867962590341477,
|
| 2270 |
+
"loss": 3.363053894042969,
|
| 2271 |
+
"step": 28700
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.576,
|
| 2275 |
+
"grad_norm": 0.15002867579460144,
|
| 2276 |
+
"learning_rate": 0.00038372930580214595,
|
| 2277 |
+
"loss": 3.3725015258789064,
|
| 2278 |
+
"step": 28800
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.578,
|
| 2282 |
+
"grad_norm": 0.14523345232009888,
|
| 2283 |
+
"learning_rate": 0.000380666970403412,
|
| 2284 |
+
"loss": 3.375093688964844,
|
| 2285 |
+
"step": 28900
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.58,
|
| 2289 |
+
"grad_norm": 0.1521645337343216,
|
| 2290 |
+
"learning_rate": 0.000377609374462353,
|
| 2291 |
+
"loss": 3.350813903808594,
|
| 2292 |
+
"step": 29000
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.58,
|
| 2296 |
+
"eval_accuracy": 0.3793452847252259,
|
| 2297 |
+
"eval_loss": 3.322114944458008,
|
| 2298 |
+
"eval_runtime": 6.9103,
|
| 2299 |
+
"eval_samples_per_second": 280.884,
|
| 2300 |
+
"eval_steps_per_second": 17.655,
|
| 2301 |
+
"step": 29000
|
| 2302 |
+
},
|
| 2303 |
+
{
|
| 2304 |
+
"epoch": 0.582,
|
| 2305 |
+
"grad_norm": 0.14382348954677582,
|
| 2306 |
+
"learning_rate": 0.00037455663941514335,
|
| 2307 |
+
"loss": 3.3781756591796874,
|
| 2308 |
+
"step": 29100
|
| 2309 |
+
},
|
| 2310 |
+
{
|
| 2311 |
+
"epoch": 0.584,
|
| 2312 |
+
"grad_norm": 0.1457841694355011,
|
| 2313 |
+
"learning_rate": 0.00037150888650490057,
|
| 2314 |
+
"loss": 3.3773403930664063,
|
| 2315 |
+
"step": 29200
|
| 2316 |
+
},
|
| 2317 |
+
{
|
| 2318 |
+
"epoch": 0.586,
|
| 2319 |
+
"grad_norm": 0.13661234080791473,
|
| 2320 |
+
"learning_rate": 0.0003684662367768703,
|
| 2321 |
+
"loss": 3.3766348266601565,
|
| 2322 |
+
"step": 29300
|
| 2323 |
+
},
|
| 2324 |
+
{
|
| 2325 |
+
"epoch": 0.588,
|
| 2326 |
+
"grad_norm": 0.16734236478805542,
|
| 2327 |
+
"learning_rate": 0.00036542881107361983,
|
| 2328 |
+
"loss": 3.3571905517578124,
|
| 2329 |
+
"step": 29400
|
| 2330 |
+
},
|
| 2331 |
+
{
|
| 2332 |
+
"epoch": 0.59,
|
| 2333 |
+
"grad_norm": 0.14157910645008087,
|
| 2334 |
+
"learning_rate": 0.00036239673003023745,
|
| 2335 |
+
"loss": 3.366629638671875,
|
| 2336 |
+
"step": 29500
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 0.592,
|
| 2340 |
+
"grad_norm": 0.13826872408390045,
|
| 2341 |
+
"learning_rate": 0.0003593701140695413,
|
| 2342 |
+
"loss": 3.3789642333984373,
|
| 2343 |
+
"step": 29600
|
| 2344 |
+
},
|
| 2345 |
+
{
|
| 2346 |
+
"epoch": 0.594,
|
| 2347 |
+
"grad_norm": 0.1601548194885254,
|
| 2348 |
+
"learning_rate": 0.00035634908339729813,
|
| 2349 |
+
"loss": 3.37031005859375,
|
| 2350 |
+
"step": 29700
|
| 2351 |
+
},
|
| 2352 |
+
{
|
| 2353 |
+
"epoch": 0.596,
|
| 2354 |
+
"grad_norm": 0.13469478487968445,
|
| 2355 |
+
"learning_rate": 0.00035333375799744704,
|
| 2356 |
+
"loss": 3.3534210205078123,
|
| 2357 |
+
"step": 29800
|
| 2358 |
+
},
|
| 2359 |
+
{
|
| 2360 |
+
"epoch": 0.598,
|
| 2361 |
+
"grad_norm": 0.12652449309825897,
|
| 2362 |
+
"learning_rate": 0.00035032425762733554,
|
| 2363 |
+
"loss": 3.34040283203125,
|
| 2364 |
+
"step": 29900
|
| 2365 |
+
},
|
| 2366 |
+
{
|
| 2367 |
+
"epoch": 0.6,
|
| 2368 |
+
"grad_norm": 0.18712250888347626,
|
| 2369 |
+
"learning_rate": 0.0003473207018129635,
|
| 2370 |
+
"loss": 3.356614685058594,
|
| 2371 |
+
"step": 30000
|
| 2372 |
+
},
|
| 2373 |
+
{
|
| 2374 |
+
"epoch": 0.6,
|
| 2375 |
+
"eval_accuracy": 0.3802063011480555,
|
| 2376 |
+
"eval_loss": 3.314772367477417,
|
| 2377 |
+
"eval_runtime": 6.363,
|
| 2378 |
+
"eval_samples_per_second": 305.046,
|
| 2379 |
+
"eval_steps_per_second": 19.173,
|
| 2380 |
+
"step": 30000
|
| 2381 |
+
},
|
| 2382 |
+
{
|
| 2383 |
+
"epoch": 0.602,
|
| 2384 |
+
"grad_norm": 0.14657410979270935,
|
| 2385 |
+
"learning_rate": 0.0003443232098442347,
|
| 2386 |
+
"loss": 3.3668572998046873,
|
| 2387 |
+
"step": 30100
|
| 2388 |
+
},
|
| 2389 |
+
{
|
| 2390 |
+
"epoch": 0.604,
|
| 2391 |
+
"grad_norm": 0.138699010014534,
|
| 2392 |
+
"learning_rate": 0.0003413319007702201,
|
| 2393 |
+
"loss": 3.3502297973632813,
|
| 2394 |
+
"step": 30200
|
| 2395 |
+
},
|
| 2396 |
+
{
|
| 2397 |
+
"epoch": 0.606,
|
| 2398 |
+
"grad_norm": 0.15832597017288208,
|
| 2399 |
+
"learning_rate": 0.0003383468933944299,
|
| 2400 |
+
"loss": 3.355211181640625,
|
| 2401 |
+
"step": 30300
|
| 2402 |
+
},
|
| 2403 |
+
{
|
| 2404 |
+
"epoch": 0.608,
|
| 2405 |
+
"grad_norm": 0.15029780566692352,
|
| 2406 |
+
"learning_rate": 0.0003353683062700939,
|
| 2407 |
+
"loss": 3.348283386230469,
|
| 2408 |
+
"step": 30400
|
| 2409 |
+
},
|
| 2410 |
+
{
|
| 2411 |
+
"epoch": 0.61,
|
| 2412 |
+
"grad_norm": 0.17595185339450836,
|
| 2413 |
+
"learning_rate": 0.0003323962576954539,
|
| 2414 |
+
"loss": 3.3513107299804688,
|
| 2415 |
+
"step": 30500
|
| 2416 |
+
},
|
| 2417 |
+
{
|
| 2418 |
+
"epoch": 0.612,
|
| 2419 |
+
"grad_norm": 0.1754045933485031,
|
| 2420 |
+
"learning_rate": 0.00032943086570906576,
|
| 2421 |
+
"loss": 3.3677444458007812,
|
| 2422 |
+
"step": 30600
|
| 2423 |
+
},
|
| 2424 |
+
{
|
| 2425 |
+
"epoch": 0.614,
|
| 2426 |
+
"grad_norm": 0.14994262158870697,
|
| 2427 |
+
"learning_rate": 0.00032647224808511,
|
| 2428 |
+
"loss": 3.3526513671875,
|
| 2429 |
+
"step": 30700
|
| 2430 |
+
},
|
| 2431 |
+
{
|
| 2432 |
+
"epoch": 0.616,
|
| 2433 |
+
"grad_norm": 0.14404474198818207,
|
| 2434 |
+
"learning_rate": 0.00032352052232871543,
|
| 2435 |
+
"loss": 3.31786865234375,
|
| 2436 |
+
"step": 30800
|
| 2437 |
+
},
|
| 2438 |
+
{
|
| 2439 |
+
"epoch": 0.618,
|
| 2440 |
+
"grad_norm": 0.14154170453548431,
|
| 2441 |
+
"learning_rate": 0.0003205758056712919,
|
| 2442 |
+
"loss": 3.341680908203125,
|
| 2443 |
+
"step": 30900
|
| 2444 |
+
},
|
| 2445 |
+
{
|
| 2446 |
+
"epoch": 0.62,
|
| 2447 |
+
"grad_norm": 0.13272085785865784,
|
| 2448 |
+
"learning_rate": 0.0003176382150658742,
|
| 2449 |
+
"loss": 3.344057922363281,
|
| 2450 |
+
"step": 31000
|
| 2451 |
+
},
|
| 2452 |
+
{
|
| 2453 |
+
"epoch": 0.62,
|
| 2454 |
+
"eval_accuracy": 0.3808566004369608,
|
| 2455 |
+
"eval_loss": 3.3054208755493164,
|
| 2456 |
+
"eval_runtime": 6.7684,
|
| 2457 |
+
"eval_samples_per_second": 286.774,
|
| 2458 |
+
"eval_steps_per_second": 18.025,
|
| 2459 |
+
"step": 31000
|
| 2460 |
+
},
|
| 2461 |
+
{
|
| 2462 |
+
"epoch": 0.622,
|
| 2463 |
+
"grad_norm": 0.14439266920089722,
|
| 2464 |
+
"learning_rate": 0.00031470786718247704,
|
| 2465 |
+
"loss": 3.336947021484375,
|
| 2466 |
+
"step": 31100
|
| 2467 |
+
},
|
| 2468 |
+
{
|
| 2469 |
+
"epoch": 0.624,
|
| 2470 |
+
"grad_norm": 0.1481838971376419,
|
| 2471 |
+
"learning_rate": 0.000311784878403462,
|
| 2472 |
+
"loss": 3.3591497802734374,
|
| 2473 |
+
"step": 31200
|
| 2474 |
+
},
|
| 2475 |
+
{
|
| 2476 |
+
"epoch": 0.626,
|
| 2477 |
+
"grad_norm": 0.13957072794437408,
|
| 2478 |
+
"learning_rate": 0.00030886936481891447,
|
| 2479 |
+
"loss": 3.333016357421875,
|
| 2480 |
+
"step": 31300
|
| 2481 |
+
},
|
| 2482 |
+
{
|
| 2483 |
+
"epoch": 0.628,
|
| 2484 |
+
"grad_norm": 0.1470046043395996,
|
| 2485 |
+
"learning_rate": 0.0003059614422220331,
|
| 2486 |
+
"loss": 3.349130859375,
|
| 2487 |
+
"step": 31400
|
| 2488 |
+
},
|
| 2489 |
+
{
|
| 2490 |
+
"epoch": 0.63,
|
| 2491 |
+
"grad_norm": 0.13627968728542328,
|
| 2492 |
+
"learning_rate": 0.00030306122610453183,
|
| 2493 |
+
"loss": 3.3334747314453126,
|
| 2494 |
+
"step": 31500
|
| 2495 |
+
},
|
| 2496 |
+
{
|
| 2497 |
+
"epoch": 0.632,
|
| 2498 |
+
"grad_norm": 0.27877408266067505,
|
| 2499 |
+
"learning_rate": 0.00030016883165205166,
|
| 2500 |
+
"loss": 3.3201373291015623,
|
| 2501 |
+
"step": 31600
|
| 2502 |
+
},
|
| 2503 |
+
{
|
| 2504 |
+
"epoch": 0.634,
|
| 2505 |
+
"grad_norm": 0.14691542088985443,
|
| 2506 |
+
"learning_rate": 0.00029728437373958684,
|
| 2507 |
+
"loss": 3.3420004272460937,
|
| 2508 |
+
"step": 31700
|
| 2509 |
+
},
|
| 2510 |
+
{
|
| 2511 |
+
"epoch": 0.636,
|
| 2512 |
+
"grad_norm": 0.1655166745185852,
|
| 2513 |
+
"learning_rate": 0.0002944079669269224,
|
| 2514 |
+
"loss": 3.32868408203125,
|
| 2515 |
+
"step": 31800
|
| 2516 |
+
},
|
| 2517 |
+
{
|
| 2518 |
+
"epoch": 0.638,
|
| 2519 |
+
"grad_norm": 0.16263002157211304,
|
| 2520 |
+
"learning_rate": 0.0002915397254540836,
|
| 2521 |
+
"loss": 3.327845153808594,
|
| 2522 |
+
"step": 31900
|
| 2523 |
+
},
|
| 2524 |
+
{
|
| 2525 |
+
"epoch": 0.64,
|
| 2526 |
+
"grad_norm": 0.13551194965839386,
|
| 2527 |
+
"learning_rate": 0.00028867976323679957,
|
| 2528 |
+
"loss": 3.326793518066406,
|
| 2529 |
+
"step": 32000
|
| 2530 |
+
},
|
| 2531 |
+
{
|
| 2532 |
+
"epoch": 0.64,
|
| 2533 |
+
"eval_accuracy": 0.38135970019690457,
|
| 2534 |
+
"eval_loss": 3.302846908569336,
|
| 2535 |
+
"eval_runtime": 6.3825,
|
| 2536 |
+
"eval_samples_per_second": 304.112,
|
| 2537 |
+
"eval_steps_per_second": 19.115,
|
| 2538 |
+
"step": 32000
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 0.002,
|
| 2542 |
+
"grad_norm": 0.14640577137470245,
|
| 2543 |
+
"learning_rate": 0.0002858281938619785,
|
| 2544 |
+
"loss": 3.115254211425781,
|
| 2545 |
+
"step": 32100
|
| 2546 |
+
},
|
| 2547 |
+
{
|
| 2548 |
+
"epoch": 0.004,
|
| 2549 |
+
"grad_norm": 0.139010488986969,
|
| 2550 |
+
"learning_rate": 0.00028298513058319685,
|
| 2551 |
+
"loss": 3.07729248046875,
|
| 2552 |
+
"step": 32200
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 0.006,
|
| 2556 |
+
"grad_norm": 0.1498369425535202,
|
| 2557 |
+
"learning_rate": 0.0002801506863162001,
|
| 2558 |
+
"loss": 3.0912158203125,
|
| 2559 |
+
"step": 32300
|
| 2560 |
+
},
|
| 2561 |
+
{
|
| 2562 |
+
"epoch": 0.008,
|
| 2563 |
+
"grad_norm": 0.15137332677841187,
|
| 2564 |
+
"learning_rate": 0.0002773249736344206,
|
| 2565 |
+
"loss": 3.07073486328125,
|
| 2566 |
+
"step": 32400
|
| 2567 |
+
},
|
| 2568 |
+
{
|
| 2569 |
+
"epoch": 0.01,
|
| 2570 |
+
"grad_norm": 0.14690713584423065,
|
| 2571 |
+
"learning_rate": 0.0002745081047645037,
|
| 2572 |
+
"loss": 3.0815118408203124,
|
| 2573 |
+
"step": 32500
|
| 2574 |
+
},
|
| 2575 |
+
{
|
| 2576 |
+
"epoch": 0.012,
|
| 2577 |
+
"grad_norm": 0.1449531465768814,
|
| 2578 |
+
"learning_rate": 0.00027170019158185253,
|
| 2579 |
+
"loss": 3.067219543457031,
|
| 2580 |
+
"step": 32600
|
| 2581 |
+
},
|
| 2582 |
+
{
|
| 2583 |
+
"epoch": 0.014,
|
| 2584 |
+
"grad_norm": 0.15289907157421112,
|
| 2585 |
+
"learning_rate": 0.0002689013456061843,
|
| 2586 |
+
"loss": 3.065643310546875,
|
| 2587 |
+
"step": 32700
|
| 2588 |
+
},
|
| 2589 |
+
{
|
| 2590 |
+
"epoch": 0.016,
|
| 2591 |
+
"grad_norm": 0.13553819060325623,
|
| 2592 |
+
"learning_rate": 0.00026611167799710044,
|
| 2593 |
+
"loss": 3.081978759765625,
|
| 2594 |
+
"step": 32800
|
| 2595 |
+
},
|
| 2596 |
+
{
|
| 2597 |
+
"epoch": 0.018,
|
| 2598 |
+
"grad_norm": 0.15303637087345123,
|
| 2599 |
+
"learning_rate": 0.0002633312995496727,
|
| 2600 |
+
"loss": 3.074805908203125,
|
| 2601 |
+
"step": 32900
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 0.02,
|
| 2605 |
+
"grad_norm": 0.1398230940103531,
|
| 2606 |
+
"learning_rate": 0.0002605603206900423,
|
| 2607 |
+
"loss": 3.061953125,
|
| 2608 |
+
"step": 33000
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 0.02,
|
| 2612 |
+
"eval_accuracy": 0.3804281086574496,
|
| 2613 |
+
"eval_loss": 3.310852289199829,
|
| 2614 |
+
"eval_runtime": 11.9581,
|
| 2615 |
+
"eval_samples_per_second": 162.316,
|
| 2616 |
+
"eval_steps_per_second": 10.202,
|
| 2617 |
+
"step": 33000
|
| 2618 |
+
},
|
| 2619 |
+
{
|
| 2620 |
+
"epoch": 0.022,
|
| 2621 |
+
"grad_norm": 0.19870808720588684,
|
| 2622 |
+
"learning_rate": 0.00025779885147103436,
|
| 2623 |
+
"loss": 3.070841979980469,
|
| 2624 |
+
"step": 33100
|
| 2625 |
+
},
|
| 2626 |
+
{
|
| 2627 |
+
"epoch": 0.024,
|
| 2628 |
+
"grad_norm": 0.15535953640937805,
|
| 2629 |
+
"learning_rate": 0.0002550470015677863,
|
| 2630 |
+
"loss": 3.064533996582031,
|
| 2631 |
+
"step": 33200
|
| 2632 |
+
},
|
| 2633 |
+
{
|
| 2634 |
+
"epoch": 0.026,
|
| 2635 |
+
"grad_norm": 0.15493761003017426,
|
| 2636 |
+
"learning_rate": 0.0002523048802733939,
|
| 2637 |
+
"loss": 3.0495779418945315,
|
| 2638 |
+
"step": 33300
|
| 2639 |
+
},
|
| 2640 |
+
{
|
| 2641 |
+
"epoch": 0.028,
|
| 2642 |
+
"grad_norm": 0.13659609854221344,
|
| 2643 |
+
"learning_rate": 0.0002495725964945681,
|
| 2644 |
+
"loss": 3.042427978515625,
|
| 2645 |
+
"step": 33400
|
| 2646 |
+
},
|
| 2647 |
+
{
|
| 2648 |
+
"epoch": 0.03,
|
| 2649 |
+
"grad_norm": 0.1522352397441864,
|
| 2650 |
+
"learning_rate": 0.0002468502587473111,
|
| 2651 |
+
"loss": 3.0562362670898438,
|
| 2652 |
+
"step": 33500
|
| 2653 |
+
},
|
| 2654 |
+
{
|
| 2655 |
+
"epoch": 0.032,
|
| 2656 |
+
"grad_norm": 0.19474385678768158,
|
| 2657 |
+
"learning_rate": 0.00024413797515260717,
|
| 2658 |
+
"loss": 3.051845703125,
|
| 2659 |
+
"step": 33600
|
| 2660 |
+
},
|
| 2661 |
+
{
|
| 2662 |
+
"epoch": 0.034,
|
| 2663 |
+
"grad_norm": 0.15455342829227448,
|
| 2664 |
+
"learning_rate": 0.0002414358534321262,
|
| 2665 |
+
"loss": 3.0544952392578124,
|
| 2666 |
+
"step": 33700
|
| 2667 |
+
},
|
| 2668 |
+
{
|
| 2669 |
+
"epoch": 0.036,
|
| 2670 |
+
"grad_norm": 0.17269164323806763,
|
| 2671 |
+
"learning_rate": 0.00023874400090394778,
|
| 2672 |
+
"loss": 3.0642709350585937,
|
| 2673 |
+
"step": 33800
|
| 2674 |
+
},
|
| 2675 |
+
{
|
| 2676 |
+
"epoch": 0.038,
|
| 2677 |
+
"grad_norm": 0.14971676468849182,
|
| 2678 |
+
"learning_rate": 0.0002360625244782974,
|
| 2679 |
+
"loss": 3.0341754150390625,
|
| 2680 |
+
"step": 33900
|
| 2681 |
+
},
|
| 2682 |
+
{
|
| 2683 |
+
"epoch": 0.04,
|
| 2684 |
+
"grad_norm": 0.14853248000144958,
|
| 2685 |
+
"learning_rate": 0.0002333915306533012,
|
| 2686 |
+
"loss": 3.015852355957031,
|
| 2687 |
+
"step": 34000
|
| 2688 |
+
},
|
| 2689 |
+
{
|
| 2690 |
+
"epoch": 0.04,
|
| 2691 |
+
"eval_accuracy": 0.38098766851069366,
|
| 2692 |
+
"eval_loss": 3.3074276447296143,
|
| 2693 |
+
"eval_runtime": 13.1242,
|
| 2694 |
+
"eval_samples_per_second": 147.895,
|
| 2695 |
+
"eval_steps_per_second": 9.296,
|
| 2696 |
+
"step": 34000
|
| 2697 |
+
}
|
| 2698 |
+
],
|
| 2699 |
+
"logging_steps": 100,
|
| 2700 |
+
"max_steps": 50000,
|
| 2701 |
+
"num_input_tokens_seen": 0,
|
| 2702 |
+
"num_train_epochs": 9223372036854775807,
|
| 2703 |
+
"save_steps": 2000,
|
| 2704 |
+
"stateful_callbacks": {
|
| 2705 |
+
"TrainerControl": {
|
| 2706 |
+
"args": {
|
| 2707 |
+
"should_epoch_stop": false,
|
| 2708 |
+
"should_evaluate": false,
|
| 2709 |
+
"should_log": false,
|
| 2710 |
+
"should_save": true,
|
| 2711 |
+
"should_training_stop": false
|
| 2712 |
+
},
|
| 2713 |
+
"attributes": {}
|
| 2714 |
+
}
|
| 2715 |
+
},
|
| 2716 |
+
"total_flos": 1.33278669471744e+18,
|
| 2717 |
+
"train_batch_size": 120,
|
| 2718 |
+
"trial_name": null,
|
| 2719 |
+
"trial_params": null
|
| 2720 |
+
}
|
last-checkpoint/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:651f72844ade54f1175bd507f2f69a0c1cdeae4fb980af8775a3f4647e063e4d
|
| 3 |
+
size 5265
|