Auto-upload checkpoint step 1800
Browse files- checkpoint-1800/config.json +17 -0
- checkpoint-1800/generation_config.json +6 -0
- checkpoint-1800/model.safetensors +3 -0
- checkpoint-1800/optimizer.pt +3 -0
- checkpoint-1800/rng_state.pth +3 -0
- checkpoint-1800/scaler.pt +3 -0
- checkpoint-1800/scheduler.pt +3 -0
- checkpoint-1800/trainer_state.json +1294 -0
- checkpoint-1800/training_args.bin +3 -0
- checkpoint-1800/vocab_multisense_8k.json +0 -0
checkpoint-1800/config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"HyperBitSSMForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "float32",
|
| 6 |
+
"hidden_size": 768,
|
| 7 |
+
"hop": 48,
|
| 8 |
+
"intermediate_size": 1792,
|
| 9 |
+
"lowrank": 32,
|
| 10 |
+
"model_type": "hyper_bit_ssm_98m_v55",
|
| 11 |
+
"num_attention_heads": 12,
|
| 12 |
+
"num_hidden_layers": 12,
|
| 13 |
+
"transformers_version": "5.15.0",
|
| 14 |
+
"use_cache": false,
|
| 15 |
+
"vocab_size": 8192,
|
| 16 |
+
"window": 96
|
| 17 |
+
}
|
checkpoint-1800/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"output_attentions": false,
|
| 4 |
+
"output_hidden_states": false,
|
| 5 |
+
"transformers_version": "5.15.0"
|
| 6 |
+
}
|
checkpoint-1800/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e58169919b728dc09265a4011dd6b48d40bf9d34d8a7bfc9440c4f7fd141449
|
| 3 |
+
size 393799132
|
checkpoint-1800/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4aeab441d0427aec2b59a37284d2a9edcdc450366ebfe4f3bde89af1e429a0c1
|
| 3 |
+
size 112281209
|
checkpoint-1800/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:01ff51baf7fc6cc08624694149425f5ce799caf8cc4e43d641fb4c6cd21e28fb
|
| 3 |
+
size 14645
|
checkpoint-1800/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2e3771fdd65f8d8134b9f412724f6216231835dccb19bb4a31906cda6ea61dd4
|
| 3 |
+
size 1383
|
checkpoint-1800/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f4be0e5c2c45537a8e259d2f662857831f1d6d503bee06069f1da7ae0eac6f61
|
| 3 |
+
size 1465
|
checkpoint-1800/trainer_state.json
ADDED
|
@@ -0,0 +1,1294 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.010517520436126514,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1800,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 5.8430669089591743e-05,
|
| 14 |
+
"grad_norm": 10.995234489440918,
|
| 15 |
+
"learning_rate": 6e-06,
|
| 16 |
+
"loss": 36.76142578125,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.00011686133817918349,
|
| 21 |
+
"grad_norm": 10.91879653930664,
|
| 22 |
+
"learning_rate": 1.2666666666666668e-05,
|
| 23 |
+
"loss": 36.63887023925781,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.00017529200726877522,
|
| 28 |
+
"grad_norm": 12.004676818847656,
|
| 29 |
+
"learning_rate": 1.9333333333333333e-05,
|
| 30 |
+
"loss": 36.38578491210937,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.00023372267635836697,
|
| 35 |
+
"grad_norm": 11.3691987991333,
|
| 36 |
+
"learning_rate": 2.6000000000000002e-05,
|
| 37 |
+
"loss": 36.06534118652344,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0002921533454479587,
|
| 42 |
+
"grad_norm": 13.08530044555664,
|
| 43 |
+
"learning_rate": 3.266666666666667e-05,
|
| 44 |
+
"loss": 35.64937744140625,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.00035058401453755045,
|
| 49 |
+
"grad_norm": 13.18602180480957,
|
| 50 |
+
"learning_rate": 3.933333333333333e-05,
|
| 51 |
+
"loss": 35.051446533203126,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.0004090146836271422,
|
| 56 |
+
"grad_norm": 13.574982643127441,
|
| 57 |
+
"learning_rate": 4.600000000000001e-05,
|
| 58 |
+
"loss": 34.180169677734376,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.00046744535271673395,
|
| 63 |
+
"grad_norm": 14.570645332336426,
|
| 64 |
+
"learning_rate": 5.266666666666666e-05,
|
| 65 |
+
"loss": 33.2185546875,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.0005258760218063257,
|
| 70 |
+
"grad_norm": 16.55890464782715,
|
| 71 |
+
"learning_rate": 5.9333333333333343e-05,
|
| 72 |
+
"loss": 32.12116394042969,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.0005843066908959174,
|
| 77 |
+
"grad_norm": 17.331336975097656,
|
| 78 |
+
"learning_rate": 6.6e-05,
|
| 79 |
+
"loss": 30.47571716308594,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.0006427373599855092,
|
| 84 |
+
"grad_norm": 14.68326187133789,
|
| 85 |
+
"learning_rate": 7.266666666666667e-05,
|
| 86 |
+
"loss": 28.728692626953126,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.0007011680290751009,
|
| 91 |
+
"grad_norm": 18.396814346313477,
|
| 92 |
+
"learning_rate": 7.933333333333334e-05,
|
| 93 |
+
"loss": 26.857958984375,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.0007595986981646927,
|
| 98 |
+
"grad_norm": 16.36812973022461,
|
| 99 |
+
"learning_rate": 8.6e-05,
|
| 100 |
+
"loss": 25.132843017578125,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.0008180293672542844,
|
| 105 |
+
"grad_norm": 21.44991111755371,
|
| 106 |
+
"learning_rate": 9.266666666666666e-05,
|
| 107 |
+
"loss": 23.530233764648436,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.0008764600363438762,
|
| 112 |
+
"grad_norm": 22.47516632080078,
|
| 113 |
+
"learning_rate": 9.933333333333334e-05,
|
| 114 |
+
"loss": 22.332640075683592,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.0009348907054334679,
|
| 119 |
+
"grad_norm": 22.053213119506836,
|
| 120 |
+
"learning_rate": 0.00010600000000000002,
|
| 121 |
+
"loss": 21.743748474121094,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.0009933213745230596,
|
| 126 |
+
"grad_norm": 26.53052520751953,
|
| 127 |
+
"learning_rate": 0.00011266666666666668,
|
| 128 |
+
"loss": 20.7405029296875,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.0010517520436126514,
|
| 133 |
+
"grad_norm": 37.71436309814453,
|
| 134 |
+
"learning_rate": 0.00011933333333333334,
|
| 135 |
+
"loss": 19.994790649414064,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.001110182712702243,
|
| 140 |
+
"grad_norm": 33.68248748779297,
|
| 141 |
+
"learning_rate": 0.000126,
|
| 142 |
+
"loss": 19.59712371826172,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.001168613381791835,
|
| 147 |
+
"grad_norm": 29.311405181884766,
|
| 148 |
+
"learning_rate": 0.00013266666666666667,
|
| 149 |
+
"loss": 19.54248809814453,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.0012270440508814267,
|
| 154 |
+
"grad_norm": 27.571697235107422,
|
| 155 |
+
"learning_rate": 0.00013933333333333335,
|
| 156 |
+
"loss": 18.915852355957032,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.0012854747199710183,
|
| 161 |
+
"grad_norm": 50.6077880859375,
|
| 162 |
+
"learning_rate": 0.000146,
|
| 163 |
+
"loss": 18.705543518066406,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.0013439053890606102,
|
| 168 |
+
"grad_norm": 24.939199447631836,
|
| 169 |
+
"learning_rate": 0.00015266666666666667,
|
| 170 |
+
"loss": 19.156636047363282,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.0014023360581502018,
|
| 175 |
+
"grad_norm": 28.88422393798828,
|
| 176 |
+
"learning_rate": 0.00015933333333333332,
|
| 177 |
+
"loss": 17.941325378417968,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.0014607667272397936,
|
| 182 |
+
"grad_norm": 32.721168518066406,
|
| 183 |
+
"learning_rate": 0.000166,
|
| 184 |
+
"loss": 17.255938720703124,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.0015191973963293855,
|
| 189 |
+
"grad_norm": 44.036075592041016,
|
| 190 |
+
"learning_rate": 0.00017266666666666667,
|
| 191 |
+
"loss": 17.748182678222655,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.001577628065418977,
|
| 196 |
+
"grad_norm": 30.866708755493164,
|
| 197 |
+
"learning_rate": 0.00017933333333333332,
|
| 198 |
+
"loss": 17.471981811523438,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.001636058734508569,
|
| 203 |
+
"grad_norm": 58.11023712158203,
|
| 204 |
+
"learning_rate": 0.00018600000000000002,
|
| 205 |
+
"loss": 17.05304870605469,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.0016944894035981605,
|
| 210 |
+
"grad_norm": 94.0334243774414,
|
| 211 |
+
"learning_rate": 0.0001926666666666667,
|
| 212 |
+
"loss": 16.84721221923828,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.0017529200726877523,
|
| 217 |
+
"grad_norm": 75.26679992675781,
|
| 218 |
+
"learning_rate": 0.00019933333333333334,
|
| 219 |
+
"loss": 16.655764770507812,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.0018113507417773442,
|
| 224 |
+
"grad_norm": NaN,
|
| 225 |
+
"learning_rate": 0.00019999999996000412,
|
| 226 |
+
"loss": 17.352436828613282,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.0018697814108669358,
|
| 231 |
+
"grad_norm": NaN,
|
| 232 |
+
"learning_rate": 0.00019999999982174674,
|
| 233 |
+
"loss": 17.325209045410155,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.0019282120799565276,
|
| 238 |
+
"grad_norm": 421.18212890625,
|
| 239 |
+
"learning_rate": 0.00019999999958473402,
|
| 240 |
+
"loss": 17.76262512207031,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.0019866427490461192,
|
| 245 |
+
"grad_norm": Infinity,
|
| 246 |
+
"learning_rate": 0.00019999999924896604,
|
| 247 |
+
"loss": 17.62756805419922,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.002045073418135711,
|
| 252 |
+
"grad_norm": 128304.0234375,
|
| 253 |
+
"learning_rate": 0.00019999999881444277,
|
| 254 |
+
"loss": 18.896243286132812,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.002103504087225303,
|
| 259 |
+
"grad_norm": 1544019.375,
|
| 260 |
+
"learning_rate": 0.00019999999828116422,
|
| 261 |
+
"loss": 19.169711303710937,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.0021619347563148947,
|
| 266 |
+
"grad_norm": 2917.049072265625,
|
| 267 |
+
"learning_rate": 0.00019999999764913037,
|
| 268 |
+
"loss": 19.900315856933595,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.002220365425404486,
|
| 273 |
+
"grad_norm": 507.0084533691406,
|
| 274 |
+
"learning_rate": 0.00019999999691834125,
|
| 275 |
+
"loss": 20.770042419433594,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.002278796094494078,
|
| 280 |
+
"grad_norm": 6932.96435546875,
|
| 281 |
+
"learning_rate": 0.00019999999608879683,
|
| 282 |
+
"loss": 21.493869018554687,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.00233722676358367,
|
| 287 |
+
"grad_norm": 711.9380493164062,
|
| 288 |
+
"learning_rate": 0.00019999999516049713,
|
| 289 |
+
"loss": 21.287228393554688,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.0023956574326732616,
|
| 294 |
+
"grad_norm": 2512.957275390625,
|
| 295 |
+
"learning_rate": 0.00019999999413344217,
|
| 296 |
+
"loss": 20.687088012695312,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.0024540881017628535,
|
| 301 |
+
"grad_norm": 245.29962158203125,
|
| 302 |
+
"learning_rate": 0.0001999999930076319,
|
| 303 |
+
"loss": 19.545924377441406,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.002512518770852445,
|
| 308 |
+
"grad_norm": 873.497314453125,
|
| 309 |
+
"learning_rate": 0.00019999999178306637,
|
| 310 |
+
"loss": 19.846446228027343,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.0025709494399420367,
|
| 315 |
+
"grad_norm": 273.556884765625,
|
| 316 |
+
"learning_rate": 0.00019999999045974555,
|
| 317 |
+
"loss": 19.4821044921875,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.0026293801090316285,
|
| 322 |
+
"grad_norm": 105.1680679321289,
|
| 323 |
+
"learning_rate": 0.00019999998903766944,
|
| 324 |
+
"loss": 18.83043212890625,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.0026878107781212203,
|
| 329 |
+
"grad_norm": 19779.220703125,
|
| 330 |
+
"learning_rate": 0.00019999998751683804,
|
| 331 |
+
"loss": 18.68498992919922,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.002746241447210812,
|
| 336 |
+
"grad_norm": 379.18695068359375,
|
| 337 |
+
"learning_rate": 0.0001999999858972514,
|
| 338 |
+
"loss": 17.952909851074217,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.0028046721163004036,
|
| 343 |
+
"grad_norm": 982.7449340820312,
|
| 344 |
+
"learning_rate": 0.00019999998417890945,
|
| 345 |
+
"loss": 18.662619018554686,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.0028631027853899954,
|
| 350 |
+
"grad_norm": 264.763671875,
|
| 351 |
+
"learning_rate": 0.00019999998236181225,
|
| 352 |
+
"loss": 18.683988952636717,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.0029215334544795872,
|
| 357 |
+
"grad_norm": 1355.505126953125,
|
| 358 |
+
"learning_rate": 0.0001999999804459598,
|
| 359 |
+
"loss": 18.34948272705078,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.002979964123569179,
|
| 364 |
+
"grad_norm": 106.23206329345703,
|
| 365 |
+
"learning_rate": 0.00019999997843135202,
|
| 366 |
+
"loss": 18.03973388671875,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.003038394792658771,
|
| 371 |
+
"grad_norm": 135.76995849609375,
|
| 372 |
+
"learning_rate": 0.00019999997631798904,
|
| 373 |
+
"loss": 18.432786560058595,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.0030968254617483623,
|
| 378 |
+
"grad_norm": 683.4022216796875,
|
| 379 |
+
"learning_rate": 0.00019999997410587078,
|
| 380 |
+
"loss": 19.00628662109375,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.003155256130837954,
|
| 385 |
+
"grad_norm": 1766.309814453125,
|
| 386 |
+
"learning_rate": 0.00019999997179499722,
|
| 387 |
+
"loss": 18.93055877685547,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.003213686799927546,
|
| 392 |
+
"grad_norm": 12474.607421875,
|
| 393 |
+
"learning_rate": 0.0001999999693853684,
|
| 394 |
+
"loss": 19.35320129394531,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.003272117469017138,
|
| 399 |
+
"grad_norm": 14713.984375,
|
| 400 |
+
"learning_rate": 0.00019999996687698437,
|
| 401 |
+
"loss": 19.091952514648437,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.0033305481381067296,
|
| 406 |
+
"grad_norm": 37941.45703125,
|
| 407 |
+
"learning_rate": 0.00019999996426984504,
|
| 408 |
+
"loss": 18.91400604248047,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.003388978807196321,
|
| 413 |
+
"grad_norm": 40131.44140625,
|
| 414 |
+
"learning_rate": 0.0001999999615639505,
|
| 415 |
+
"loss": 19.18461456298828,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.003447409476285913,
|
| 420 |
+
"grad_norm": 4507.1064453125,
|
| 421 |
+
"learning_rate": 0.00019999995875930068,
|
| 422 |
+
"loss": 19.337628173828126,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.0035058401453755047,
|
| 427 |
+
"grad_norm": 69082.234375,
|
| 428 |
+
"learning_rate": 0.00019999995585589562,
|
| 429 |
+
"loss": 19.456033325195314,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.0035642708144650965,
|
| 434 |
+
"grad_norm": 3494.4580078125,
|
| 435 |
+
"learning_rate": 0.0001999999528537353,
|
| 436 |
+
"loss": 19.609593200683594,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.0036227014835546884,
|
| 441 |
+
"grad_norm": 66.80265808105469,
|
| 442 |
+
"learning_rate": 0.00019999994975281976,
|
| 443 |
+
"loss": 19.908061218261718,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.0036811321526442797,
|
| 448 |
+
"grad_norm": 292724.375,
|
| 449 |
+
"learning_rate": 0.00019999994655314898,
|
| 450 |
+
"loss": 19.095835876464843,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.0037395628217338716,
|
| 455 |
+
"grad_norm": 27531.40625,
|
| 456 |
+
"learning_rate": 0.00019999994325472297,
|
| 457 |
+
"loss": 18.935963439941407,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.0037979934908234634,
|
| 462 |
+
"grad_norm": 82.52088165283203,
|
| 463 |
+
"learning_rate": 0.0001999999398575417,
|
| 464 |
+
"loss": 18.101783752441406,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.0038564241599130552,
|
| 469 |
+
"grad_norm": 654071.125,
|
| 470 |
+
"learning_rate": 0.0001999999363616052,
|
| 471 |
+
"loss": 19.08836212158203,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.003914854829002647,
|
| 476 |
+
"grad_norm": 2386539.75,
|
| 477 |
+
"learning_rate": 0.00019999993276691348,
|
| 478 |
+
"loss": 18.599899291992188,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.0039732854980922385,
|
| 483 |
+
"grad_norm": 54.48868179321289,
|
| 484 |
+
"learning_rate": 0.00019999992907346656,
|
| 485 |
+
"loss": 18.787478637695312,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.00403171616718183,
|
| 490 |
+
"grad_norm": 465.9234924316406,
|
| 491 |
+
"learning_rate": 0.00019999992528126442,
|
| 492 |
+
"loss": 18.208659362792968,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.004090146836271422,
|
| 497 |
+
"grad_norm": 66.7064437866211,
|
| 498 |
+
"learning_rate": 0.00019999992139030704,
|
| 499 |
+
"loss": 18.39374542236328,
|
| 500 |
+
"step": 700
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.004148577505361014,
|
| 504 |
+
"grad_norm": 86624.9921875,
|
| 505 |
+
"learning_rate": 0.00019999991740059448,
|
| 506 |
+
"loss": 18.63762969970703,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.004207008174450606,
|
| 511 |
+
"grad_norm": 79.30855560302734,
|
| 512 |
+
"learning_rate": 0.00019999991331212669,
|
| 513 |
+
"loss": 19.043922424316406,
|
| 514 |
+
"step": 720
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.004265438843540198,
|
| 518 |
+
"grad_norm": 275867.625,
|
| 519 |
+
"learning_rate": 0.00019999990912490372,
|
| 520 |
+
"loss": 18.727554321289062,
|
| 521 |
+
"step": 730
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.0043238695126297895,
|
| 525 |
+
"grad_norm": 2455.1982421875,
|
| 526 |
+
"learning_rate": 0.00019999990483892551,
|
| 527 |
+
"loss": 18.933006286621094,
|
| 528 |
+
"step": 740
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.004382300181719381,
|
| 532 |
+
"grad_norm": 89.4070816040039,
|
| 533 |
+
"learning_rate": 0.00019999990045419213,
|
| 534 |
+
"loss": 18.374476623535156,
|
| 535 |
+
"step": 750
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.004440730850808972,
|
| 539 |
+
"grad_norm": 82.21696472167969,
|
| 540 |
+
"learning_rate": 0.00019999989597070357,
|
| 541 |
+
"loss": 18.830426025390626,
|
| 542 |
+
"step": 760
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.004499161519898564,
|
| 546 |
+
"grad_norm": 92.64002990722656,
|
| 547 |
+
"learning_rate": 0.00019999989138845984,
|
| 548 |
+
"loss": 19.133992004394532,
|
| 549 |
+
"step": 770
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.004557592188988156,
|
| 553 |
+
"grad_norm": 71.72074127197266,
|
| 554 |
+
"learning_rate": 0.00019999988670746087,
|
| 555 |
+
"loss": 18.759793090820313,
|
| 556 |
+
"step": 780
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.004616022858077748,
|
| 560 |
+
"grad_norm": 69.16524505615234,
|
| 561 |
+
"learning_rate": 0.00019999988192770677,
|
| 562 |
+
"loss": 18.654483032226562,
|
| 563 |
+
"step": 790
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.00467445352716734,
|
| 567 |
+
"grad_norm": NaN,
|
| 568 |
+
"learning_rate": 0.0001999998770491975,
|
| 569 |
+
"loss": 18.783712768554686,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.004732884196256931,
|
| 574 |
+
"grad_norm": 74.00431823730469,
|
| 575 |
+
"learning_rate": 0.00019999987207193305,
|
| 576 |
+
"loss": 19.0465087890625,
|
| 577 |
+
"step": 810
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.004791314865346523,
|
| 581 |
+
"grad_norm": 82.20348358154297,
|
| 582 |
+
"learning_rate": 0.00019999986699591345,
|
| 583 |
+
"loss": 18.51463928222656,
|
| 584 |
+
"step": 820
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.004849745534436115,
|
| 588 |
+
"grad_norm": 74.8928451538086,
|
| 589 |
+
"learning_rate": 0.00019999986182113867,
|
| 590 |
+
"loss": 19.12200469970703,
|
| 591 |
+
"step": 830
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.004908176203525707,
|
| 595 |
+
"grad_norm": 99.61798858642578,
|
| 596 |
+
"learning_rate": 0.00019999985654760877,
|
| 597 |
+
"loss": 18.73216857910156,
|
| 598 |
+
"step": 840
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.004966606872615299,
|
| 602 |
+
"grad_norm": 67.76683807373047,
|
| 603 |
+
"learning_rate": 0.00019999985117532369,
|
| 604 |
+
"loss": 18.63970489501953,
|
| 605 |
+
"step": 850
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.00502503754170489,
|
| 609 |
+
"grad_norm": 85.94335174560547,
|
| 610 |
+
"learning_rate": 0.00019999984570428348,
|
| 611 |
+
"loss": 18.420225524902342,
|
| 612 |
+
"step": 860
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.0050834682107944815,
|
| 616 |
+
"grad_norm": 56.784420013427734,
|
| 617 |
+
"learning_rate": 0.00019999984013448815,
|
| 618 |
+
"loss": 18.97998962402344,
|
| 619 |
+
"step": 870
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.005141898879884073,
|
| 623 |
+
"grad_norm": 78.67700958251953,
|
| 624 |
+
"learning_rate": 0.00019999983446593767,
|
| 625 |
+
"loss": 18.795469665527342,
|
| 626 |
+
"step": 880
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.005200329548973665,
|
| 630 |
+
"grad_norm": 256.23828125,
|
| 631 |
+
"learning_rate": 0.00019999982869863206,
|
| 632 |
+
"loss": 18.619613647460938,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.005258760218063257,
|
| 637 |
+
"grad_norm": 67.97850036621094,
|
| 638 |
+
"learning_rate": 0.00019999982283257136,
|
| 639 |
+
"loss": 18.76555480957031,
|
| 640 |
+
"step": 900
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.005317190887152849,
|
| 644 |
+
"grad_norm": 168.03530883789062,
|
| 645 |
+
"learning_rate": 0.00019999981686775553,
|
| 646 |
+
"loss": 17.73389587402344,
|
| 647 |
+
"step": 910
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.005375621556242441,
|
| 651 |
+
"grad_norm": 68.2422866821289,
|
| 652 |
+
"learning_rate": 0.00019999981080418458,
|
| 653 |
+
"loss": 18.617031860351563,
|
| 654 |
+
"step": 920
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.0054340522253320325,
|
| 658 |
+
"grad_norm": 57.73808288574219,
|
| 659 |
+
"learning_rate": 0.00019999980464185853,
|
| 660 |
+
"loss": 19.143682861328124,
|
| 661 |
+
"step": 930
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.005492482894421624,
|
| 665 |
+
"grad_norm": 62.3519287109375,
|
| 666 |
+
"learning_rate": 0.0001999997983807774,
|
| 667 |
+
"loss": 18.458053588867188,
|
| 668 |
+
"step": 940
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.005550913563511216,
|
| 672 |
+
"grad_norm": 66.54176330566406,
|
| 673 |
+
"learning_rate": 0.0001999997920209412,
|
| 674 |
+
"loss": 18.606898498535156,
|
| 675 |
+
"step": 950
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.005609344232600807,
|
| 679 |
+
"grad_norm": 875.6220703125,
|
| 680 |
+
"learning_rate": 0.00019999978556234986,
|
| 681 |
+
"loss": 18.866278076171874,
|
| 682 |
+
"step": 960
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.005667774901690399,
|
| 686 |
+
"grad_norm": 126.01384735107422,
|
| 687 |
+
"learning_rate": 0.00019999977900500348,
|
| 688 |
+
"loss": 18.845365905761717,
|
| 689 |
+
"step": 970
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.005726205570779991,
|
| 693 |
+
"grad_norm": 83.26707458496094,
|
| 694 |
+
"learning_rate": 0.00019999977234890203,
|
| 695 |
+
"loss": 19.808224487304688,
|
| 696 |
+
"step": 980
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.005784636239869583,
|
| 700 |
+
"grad_norm": 69.96729278564453,
|
| 701 |
+
"learning_rate": 0.00019999976559404548,
|
| 702 |
+
"loss": 19.09478302001953,
|
| 703 |
+
"step": 990
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.0058430669089591745,
|
| 707 |
+
"grad_norm": 69.78528594970703,
|
| 708 |
+
"learning_rate": 0.00019999975874043395,
|
| 709 |
+
"loss": 19.883102416992188,
|
| 710 |
+
"step": 1000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.005901497578048766,
|
| 714 |
+
"grad_norm": 70.94114685058594,
|
| 715 |
+
"learning_rate": 0.0001999997517880673,
|
| 716 |
+
"loss": 19.841079711914062,
|
| 717 |
+
"step": 1010
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.005959928247138358,
|
| 721 |
+
"grad_norm": 61.711002349853516,
|
| 722 |
+
"learning_rate": 0.00019999974473694564,
|
| 723 |
+
"loss": 19.284510803222656,
|
| 724 |
+
"step": 1020
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.00601835891622795,
|
| 728 |
+
"grad_norm": 68237.375,
|
| 729 |
+
"learning_rate": 0.00019999973758706894,
|
| 730 |
+
"loss": 20.16834716796875,
|
| 731 |
+
"step": 1030
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.006076789585317542,
|
| 735 |
+
"grad_norm": 78.14495086669922,
|
| 736 |
+
"learning_rate": 0.0001999997303384372,
|
| 737 |
+
"loss": 19.60107421875,
|
| 738 |
+
"step": 1040
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.006135220254407134,
|
| 742 |
+
"grad_norm": 70.94654846191406,
|
| 743 |
+
"learning_rate": 0.00019999972299105046,
|
| 744 |
+
"loss": 19.730165100097658,
|
| 745 |
+
"step": 1050
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.006193650923496725,
|
| 749 |
+
"grad_norm": 67.84829711914062,
|
| 750 |
+
"learning_rate": 0.0001999997155449087,
|
| 751 |
+
"loss": 20.30048522949219,
|
| 752 |
+
"step": 1060
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.006252081592586316,
|
| 756 |
+
"grad_norm": 64.95204162597656,
|
| 757 |
+
"learning_rate": 0.0001999997080000119,
|
| 758 |
+
"loss": 19.58603515625,
|
| 759 |
+
"step": 1070
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.006310512261675908,
|
| 763 |
+
"grad_norm": 312.13201904296875,
|
| 764 |
+
"learning_rate": 0.00019999970035636014,
|
| 765 |
+
"loss": 19.632809448242188,
|
| 766 |
+
"step": 1080
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.0063689429307655,
|
| 770 |
+
"grad_norm": 74.49839782714844,
|
| 771 |
+
"learning_rate": 0.00019999969261395335,
|
| 772 |
+
"loss": 19.465110778808594,
|
| 773 |
+
"step": 1090
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.006427373599855092,
|
| 777 |
+
"grad_norm": 79.05039978027344,
|
| 778 |
+
"learning_rate": 0.0001999996847727916,
|
| 779 |
+
"loss": 20.55304718017578,
|
| 780 |
+
"step": 1100
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.006485804268944684,
|
| 784 |
+
"grad_norm": 108.85382843017578,
|
| 785 |
+
"learning_rate": 0.00019999967683287487,
|
| 786 |
+
"loss": 20.197222900390624,
|
| 787 |
+
"step": 1110
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.006544234938034276,
|
| 791 |
+
"grad_norm": 83.11217498779297,
|
| 792 |
+
"learning_rate": 0.00019999966879420317,
|
| 793 |
+
"loss": 20.842535400390624,
|
| 794 |
+
"step": 1120
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.006602665607123867,
|
| 798 |
+
"grad_norm": 84.1980209350586,
|
| 799 |
+
"learning_rate": 0.0001999996606567765,
|
| 800 |
+
"loss": 20.543856811523437,
|
| 801 |
+
"step": 1130
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.006661096276213459,
|
| 805 |
+
"grad_norm": 80.23883819580078,
|
| 806 |
+
"learning_rate": 0.0001999996524205949,
|
| 807 |
+
"loss": 20.87583770751953,
|
| 808 |
+
"step": 1140
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.006719526945303051,
|
| 812 |
+
"grad_norm": 87.83684539794922,
|
| 813 |
+
"learning_rate": 0.00019999964408565834,
|
| 814 |
+
"loss": 20.476596069335937,
|
| 815 |
+
"step": 1150
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.006777957614392642,
|
| 819 |
+
"grad_norm": 74.14551544189453,
|
| 820 |
+
"learning_rate": 0.00019999963565196685,
|
| 821 |
+
"loss": 20.464306640625,
|
| 822 |
+
"step": 1160
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 0.006836388283482234,
|
| 826 |
+
"grad_norm": 5455.28076171875,
|
| 827 |
+
"learning_rate": 0.0001999996271195204,
|
| 828 |
+
"loss": 20.442652893066406,
|
| 829 |
+
"step": 1170
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 0.006894818952571826,
|
| 833 |
+
"grad_norm": 90.34612274169922,
|
| 834 |
+
"learning_rate": 0.00019999961848831907,
|
| 835 |
+
"loss": 21.020123291015626,
|
| 836 |
+
"step": 1180
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.0069532496216614175,
|
| 840 |
+
"grad_norm": 96.43968200683594,
|
| 841 |
+
"learning_rate": 0.00019999960975836283,
|
| 842 |
+
"loss": 20.811376953125,
|
| 843 |
+
"step": 1190
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.007011680290751009,
|
| 847 |
+
"grad_norm": 110.07880401611328,
|
| 848 |
+
"learning_rate": 0.00019999960092965166,
|
| 849 |
+
"loss": 20.375579833984375,
|
| 850 |
+
"step": 1200
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.007070110959840601,
|
| 854 |
+
"grad_norm": 100.05951690673828,
|
| 855 |
+
"learning_rate": 0.0001999995920021856,
|
| 856 |
+
"loss": 20.583438110351562,
|
| 857 |
+
"step": 1210
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 0.007128541628930193,
|
| 861 |
+
"grad_norm": 103.42561340332031,
|
| 862 |
+
"learning_rate": 0.00019999958297596467,
|
| 863 |
+
"loss": 20.88934783935547,
|
| 864 |
+
"step": 1220
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 0.007186972298019785,
|
| 868 |
+
"grad_norm": 129.6129608154297,
|
| 869 |
+
"learning_rate": 0.00019999957385098886,
|
| 870 |
+
"loss": 20.9843994140625,
|
| 871 |
+
"step": 1230
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 0.007245402967109377,
|
| 875 |
+
"grad_norm": 121.91087341308594,
|
| 876 |
+
"learning_rate": 0.00019999956462725816,
|
| 877 |
+
"loss": 21.024826049804688,
|
| 878 |
+
"step": 1240
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.0073038336361989685,
|
| 882 |
+
"grad_norm": 268.5066223144531,
|
| 883 |
+
"learning_rate": 0.00019999955530477262,
|
| 884 |
+
"loss": 22.091616821289062,
|
| 885 |
+
"step": 1250
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.0073622643052885595,
|
| 889 |
+
"grad_norm": 178.2930450439453,
|
| 890 |
+
"learning_rate": 0.00019999954588353225,
|
| 891 |
+
"loss": 21.91973876953125,
|
| 892 |
+
"step": 1260
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 0.007420694974378151,
|
| 896 |
+
"grad_norm": 196.05052185058594,
|
| 897 |
+
"learning_rate": 0.00019999953636353702,
|
| 898 |
+
"loss": 21.607716369628907,
|
| 899 |
+
"step": 1270
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 0.007479125643467743,
|
| 903 |
+
"grad_norm": 176.58526611328125,
|
| 904 |
+
"learning_rate": 0.00019999952674478694,
|
| 905 |
+
"loss": 21.24676055908203,
|
| 906 |
+
"step": 1280
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.007537556312557335,
|
| 910 |
+
"grad_norm": 146.85780334472656,
|
| 911 |
+
"learning_rate": 0.00019999951702728207,
|
| 912 |
+
"loss": 21.038546752929687,
|
| 913 |
+
"step": 1290
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 0.007595986981646927,
|
| 917 |
+
"grad_norm": 126.78016662597656,
|
| 918 |
+
"learning_rate": 0.0001999995072110224,
|
| 919 |
+
"loss": 22.327189636230468,
|
| 920 |
+
"step": 1300
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.007654417650736519,
|
| 924 |
+
"grad_norm": 123.16789245605469,
|
| 925 |
+
"learning_rate": 0.0001999994972960079,
|
| 926 |
+
"loss": 22.063505554199217,
|
| 927 |
+
"step": 1310
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.0077128483198261105,
|
| 931 |
+
"grad_norm": 158.87388610839844,
|
| 932 |
+
"learning_rate": 0.00019999948728223861,
|
| 933 |
+
"loss": 23.17200164794922,
|
| 934 |
+
"step": 1320
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 0.007771278988915702,
|
| 938 |
+
"grad_norm": 137.29638671875,
|
| 939 |
+
"learning_rate": 0.00019999947716971452,
|
| 940 |
+
"loss": 21.89136657714844,
|
| 941 |
+
"step": 1330
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 0.007829709658005293,
|
| 945 |
+
"grad_norm": 115.09381866455078,
|
| 946 |
+
"learning_rate": 0.0001999994669584357,
|
| 947 |
+
"loss": 20.484407043457033,
|
| 948 |
+
"step": 1340
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 0.007888140327094886,
|
| 952 |
+
"grad_norm": 177.02696228027344,
|
| 953 |
+
"learning_rate": 0.00019999945664840208,
|
| 954 |
+
"loss": 23.307778930664064,
|
| 955 |
+
"step": 1350
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.007946570996184477,
|
| 959 |
+
"grad_norm": 147.12759399414062,
|
| 960 |
+
"learning_rate": 0.00019999944623961372,
|
| 961 |
+
"loss": 22.527589416503908,
|
| 962 |
+
"step": 1360
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.00800500166527407,
|
| 966 |
+
"grad_norm": 137.78573608398438,
|
| 967 |
+
"learning_rate": 0.00019999943573207062,
|
| 968 |
+
"loss": 22.563661193847658,
|
| 969 |
+
"step": 1370
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.00806343233436366,
|
| 973 |
+
"grad_norm": 126.27091979980469,
|
| 974 |
+
"learning_rate": 0.00019999942512577282,
|
| 975 |
+
"loss": 23.21637878417969,
|
| 976 |
+
"step": 1380
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 0.008121863003453253,
|
| 980 |
+
"grad_norm": 123.52033233642578,
|
| 981 |
+
"learning_rate": 0.00019999941442072025,
|
| 982 |
+
"loss": 22.52232971191406,
|
| 983 |
+
"step": 1390
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 0.008180293672542844,
|
| 987 |
+
"grad_norm": 138.58897399902344,
|
| 988 |
+
"learning_rate": 0.000199999403616913,
|
| 989 |
+
"loss": 24.166043090820313,
|
| 990 |
+
"step": 1400
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 0.008238724341632435,
|
| 994 |
+
"grad_norm": 151.79246520996094,
|
| 995 |
+
"learning_rate": 0.00019999939271435105,
|
| 996 |
+
"loss": 22.107101440429688,
|
| 997 |
+
"step": 1410
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 0.008297155010722028,
|
| 1001 |
+
"grad_norm": 169.4059295654297,
|
| 1002 |
+
"learning_rate": 0.0001999993817130344,
|
| 1003 |
+
"loss": 22.441184997558594,
|
| 1004 |
+
"step": 1420
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.008355585679811619,
|
| 1008 |
+
"grad_norm": 150.4652099609375,
|
| 1009 |
+
"learning_rate": 0.00019999937061296308,
|
| 1010 |
+
"loss": 22.969837951660157,
|
| 1011 |
+
"step": 1430
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.008414016348901212,
|
| 1015 |
+
"grad_norm": 165.94554138183594,
|
| 1016 |
+
"learning_rate": 0.0001999993594141371,
|
| 1017 |
+
"loss": 22.781350708007814,
|
| 1018 |
+
"step": 1440
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.008472447017990803,
|
| 1022 |
+
"grad_norm": 140.3316192626953,
|
| 1023 |
+
"learning_rate": 0.00019999934811655646,
|
| 1024 |
+
"loss": 23.747430419921876,
|
| 1025 |
+
"step": 1450
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 0.008530877687080395,
|
| 1029 |
+
"grad_norm": 169.3747100830078,
|
| 1030 |
+
"learning_rate": 0.00019999933672022114,
|
| 1031 |
+
"loss": 22.896360778808592,
|
| 1032 |
+
"step": 1460
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 0.008589308356169986,
|
| 1036 |
+
"grad_norm": 126.21299743652344,
|
| 1037 |
+
"learning_rate": 0.00019999932522513122,
|
| 1038 |
+
"loss": 23.907861328125,
|
| 1039 |
+
"step": 1470
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 0.008647739025259579,
|
| 1043 |
+
"grad_norm": 154.08580017089844,
|
| 1044 |
+
"learning_rate": 0.00019999931363128668,
|
| 1045 |
+
"loss": 23.62522735595703,
|
| 1046 |
+
"step": 1480
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.00870616969434917,
|
| 1050 |
+
"grad_norm": 188.39280700683594,
|
| 1051 |
+
"learning_rate": 0.00019999930193868753,
|
| 1052 |
+
"loss": 23.966873168945312,
|
| 1053 |
+
"step": 1490
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.008764600363438763,
|
| 1057 |
+
"grad_norm": 215.01065063476562,
|
| 1058 |
+
"learning_rate": 0.00019999929014733378,
|
| 1059 |
+
"loss": 24.547561645507812,
|
| 1060 |
+
"step": 1500
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 0.008823031032528354,
|
| 1064 |
+
"grad_norm": 232.02037048339844,
|
| 1065 |
+
"learning_rate": 0.00019999927825722544,
|
| 1066 |
+
"loss": 24.471534729003906,
|
| 1067 |
+
"step": 1510
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 0.008881461701617945,
|
| 1071 |
+
"grad_norm": 149.41700744628906,
|
| 1072 |
+
"learning_rate": 0.00019999926626836252,
|
| 1073 |
+
"loss": 24.696409606933592,
|
| 1074 |
+
"step": 1520
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.008939892370707537,
|
| 1078 |
+
"grad_norm": 156.85385131835938,
|
| 1079 |
+
"learning_rate": 0.00019999925418074502,
|
| 1080 |
+
"loss": 24.592039489746092,
|
| 1081 |
+
"step": 1530
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 0.008998323039797128,
|
| 1085 |
+
"grad_norm": 147.45748901367188,
|
| 1086 |
+
"learning_rate": 0.00019999924199437302,
|
| 1087 |
+
"loss": 24.55860595703125,
|
| 1088 |
+
"step": 1540
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.009056753708886721,
|
| 1092 |
+
"grad_norm": 145.37066650390625,
|
| 1093 |
+
"learning_rate": 0.00019999922970924644,
|
| 1094 |
+
"loss": 25.61136474609375,
|
| 1095 |
+
"step": 1550
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.009115184377976312,
|
| 1099 |
+
"grad_norm": 135.78976440429688,
|
| 1100 |
+
"learning_rate": 0.00019999921732536535,
|
| 1101 |
+
"loss": 24.989341735839844,
|
| 1102 |
+
"step": 1560
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 0.009173615047065905,
|
| 1106 |
+
"grad_norm": 195.50225830078125,
|
| 1107 |
+
"learning_rate": 0.00019999920484272971,
|
| 1108 |
+
"loss": 24.82658996582031,
|
| 1109 |
+
"step": 1570
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.009232045716155496,
|
| 1113 |
+
"grad_norm": 154.9837188720703,
|
| 1114 |
+
"learning_rate": 0.0001999991922613396,
|
| 1115 |
+
"loss": 25.492460632324217,
|
| 1116 |
+
"step": 1580
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.009290476385245088,
|
| 1120 |
+
"grad_norm": 339.68975830078125,
|
| 1121 |
+
"learning_rate": 0.000199999179581195,
|
| 1122 |
+
"loss": 26.137579345703124,
|
| 1123 |
+
"step": 1590
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.00934890705433468,
|
| 1127 |
+
"grad_norm": 269.67041015625,
|
| 1128 |
+
"learning_rate": 0.0001999991668022959,
|
| 1129 |
+
"loss": 26.144183349609374,
|
| 1130 |
+
"step": 1600
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.00940733772342427,
|
| 1134 |
+
"grad_norm": 181.44190979003906,
|
| 1135 |
+
"learning_rate": 0.00019999915392464237,
|
| 1136 |
+
"loss": 27.237515258789063,
|
| 1137 |
+
"step": 1610
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.009465768392513863,
|
| 1141 |
+
"grad_norm": 189.5352020263672,
|
| 1142 |
+
"learning_rate": 0.00019999914094823436,
|
| 1143 |
+
"loss": 25.281341552734375,
|
| 1144 |
+
"step": 1620
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 0.009524199061603454,
|
| 1148 |
+
"grad_norm": 190.4973602294922,
|
| 1149 |
+
"learning_rate": 0.0001999991278730719,
|
| 1150 |
+
"loss": 26.445278930664063,
|
| 1151 |
+
"step": 1630
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 0.009582629730693046,
|
| 1155 |
+
"grad_norm": 293.44732666015625,
|
| 1156 |
+
"learning_rate": 0.00019999911469915503,
|
| 1157 |
+
"loss": 26.505374145507812,
|
| 1158 |
+
"step": 1640
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 0.009641060399782637,
|
| 1162 |
+
"grad_norm": 241.70323181152344,
|
| 1163 |
+
"learning_rate": 0.00019999910142648374,
|
| 1164 |
+
"loss": 27.189990234375,
|
| 1165 |
+
"step": 1650
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 0.00969949106887223,
|
| 1169 |
+
"grad_norm": 168.84898376464844,
|
| 1170 |
+
"learning_rate": 0.00019999908805505806,
|
| 1171 |
+
"loss": 25.387188720703126,
|
| 1172 |
+
"step": 1660
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.009757921737961821,
|
| 1176 |
+
"grad_norm": 160.4290313720703,
|
| 1177 |
+
"learning_rate": 0.000199999074584878,
|
| 1178 |
+
"loss": 26.115573120117187,
|
| 1179 |
+
"step": 1670
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.009816352407051414,
|
| 1183 |
+
"grad_norm": 170.6302947998047,
|
| 1184 |
+
"learning_rate": 0.00019999906101594354,
|
| 1185 |
+
"loss": 28.13587646484375,
|
| 1186 |
+
"step": 1680
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.009874783076141005,
|
| 1190 |
+
"grad_norm": 195.57412719726562,
|
| 1191 |
+
"learning_rate": 0.00019999904734825473,
|
| 1192 |
+
"loss": 28.04927978515625,
|
| 1193 |
+
"step": 1690
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 0.009933213745230597,
|
| 1197 |
+
"grad_norm": 180.4416961669922,
|
| 1198 |
+
"learning_rate": 0.0001999990335818116,
|
| 1199 |
+
"loss": 27.177688598632812,
|
| 1200 |
+
"step": 1700
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 0.009991644414320188,
|
| 1204 |
+
"grad_norm": 178.05911254882812,
|
| 1205 |
+
"learning_rate": 0.00019999901971661408,
|
| 1206 |
+
"loss": 28.898239135742188,
|
| 1207 |
+
"step": 1710
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 0.01005007508340978,
|
| 1211 |
+
"grad_norm": 168.58255004882812,
|
| 1212 |
+
"learning_rate": 0.00019999900575266228,
|
| 1213 |
+
"loss": 27.87994384765625,
|
| 1214 |
+
"step": 1720
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.010108505752499372,
|
| 1218 |
+
"grad_norm": 222.35963439941406,
|
| 1219 |
+
"learning_rate": 0.00019999899168995615,
|
| 1220 |
+
"loss": 28.254971313476563,
|
| 1221 |
+
"step": 1730
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 0.010166936421588963,
|
| 1225 |
+
"grad_norm": 174.24864196777344,
|
| 1226 |
+
"learning_rate": 0.00019999897752849574,
|
| 1227 |
+
"loss": 28.35985412597656,
|
| 1228 |
+
"step": 1740
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 0.010225367090678556,
|
| 1232 |
+
"grad_norm": 178.28189086914062,
|
| 1233 |
+
"learning_rate": 0.00019999896326828104,
|
| 1234 |
+
"loss": 29.379049682617186,
|
| 1235 |
+
"step": 1750
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 0.010283797759768147,
|
| 1239 |
+
"grad_norm": 184.63119506835938,
|
| 1240 |
+
"learning_rate": 0.0001999989489093121,
|
| 1241 |
+
"loss": 28.810888671875,
|
| 1242 |
+
"step": 1760
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.01034222842885774,
|
| 1246 |
+
"grad_norm": 169.72625732421875,
|
| 1247 |
+
"learning_rate": 0.0001999989344515889,
|
| 1248 |
+
"loss": 28.164144897460936,
|
| 1249 |
+
"step": 1770
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 0.01040065909794733,
|
| 1253 |
+
"grad_norm": 254.48243713378906,
|
| 1254 |
+
"learning_rate": 0.00019999891989511145,
|
| 1255 |
+
"loss": 30.535433959960937,
|
| 1256 |
+
"step": 1780
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.010459089767036923,
|
| 1260 |
+
"grad_norm": 186.9215850830078,
|
| 1261 |
+
"learning_rate": 0.0001999989052398798,
|
| 1262 |
+
"loss": 29.578173828125,
|
| 1263 |
+
"step": 1790
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.010517520436126514,
|
| 1267 |
+
"grad_norm": 324.0770568847656,
|
| 1268 |
+
"learning_rate": 0.00019999889048589396,
|
| 1269 |
+
"loss": 29.889486694335936,
|
| 1270 |
+
"step": 1800
|
| 1271 |
+
}
|
| 1272 |
+
],
|
| 1273 |
+
"logging_steps": 10,
|
| 1274 |
+
"max_steps": 1000000,
|
| 1275 |
+
"num_input_tokens_seen": 0,
|
| 1276 |
+
"num_train_epochs": 6,
|
| 1277 |
+
"save_steps": 200,
|
| 1278 |
+
"stateful_callbacks": {
|
| 1279 |
+
"TrainerControl": {
|
| 1280 |
+
"args": {
|
| 1281 |
+
"should_epoch_stop": false,
|
| 1282 |
+
"should_evaluate": false,
|
| 1283 |
+
"should_log": false,
|
| 1284 |
+
"should_save": true,
|
| 1285 |
+
"should_training_stop": false
|
| 1286 |
+
},
|
| 1287 |
+
"attributes": {}
|
| 1288 |
+
}
|
| 1289 |
+
},
|
| 1290 |
+
"total_flos": 1.444229702559744e+16,
|
| 1291 |
+
"train_batch_size": 8,
|
| 1292 |
+
"trial_name": null,
|
| 1293 |
+
"trial_params": null
|
| 1294 |
+
}
|
checkpoint-1800/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94dfe38cb2077b185e54e38dfc9c034320652e42ed5dcdcac0babd185872d525
|
| 3 |
+
size 5265
|
checkpoint-1800/vocab_multisense_8k.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|