Farouk
commited on
Commit
·
bb9548f
1
Parent(s):
3468485
Training in progress, step 9800
Browse files- adapter_model.bin +1 -1
- checkpoint-6200/adapter_model/adapter_model/README.md +12 -0
- checkpoint-6200/adapter_model/adapter_model/adapter_model.bin +1 -1
- {checkpoint-7800 → checkpoint-9800}/README.md +0 -0
- {checkpoint-7800 → checkpoint-9800}/adapter_config.json +0 -0
- {checkpoint-7800 → checkpoint-9800}/adapter_model.bin +1 -1
- {checkpoint-7800 → checkpoint-9800}/added_tokens.json +0 -0
- {checkpoint-7800 → checkpoint-9800}/optimizer.pt +1 -1
- {checkpoint-7800 → checkpoint-9800}/rng_state.pth +1 -1
- {checkpoint-7800 → checkpoint-9800}/scheduler.pt +1 -1
- {checkpoint-7800 → checkpoint-9800}/special_tokens_map.json +0 -0
- {checkpoint-7800 → checkpoint-9800}/tokenizer.model +0 -0
- {checkpoint-7800 → checkpoint-9800}/tokenizer_config.json +0 -0
- {checkpoint-7800 → checkpoint-9800}/trainer_state.json +1913 -3
- {checkpoint-7800 → checkpoint-9800}/training_args.bin +0 -0
adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 319977229
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:001215d79dce3fe3434e00202a8d7189e8f6c0796c7f37e48480e76aa0ae11a0
|
| 3 |
size 319977229
|
checkpoint-6200/adapter_model/adapter_model/README.md
CHANGED
|
@@ -180,6 +180,17 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 180 |
- bnb_4bit_use_double_quant: True
|
| 181 |
- bnb_4bit_compute_dtype: bfloat16
|
| 182 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 183 |
The following `bitsandbytes` quantization config was used during training:
|
| 184 |
- load_in_8bit: False
|
| 185 |
- load_in_4bit: True
|
|
@@ -208,5 +219,6 @@ The following `bitsandbytes` quantization config was used during training:
|
|
| 208 |
- PEFT 0.4.0
|
| 209 |
- PEFT 0.4.0
|
| 210 |
- PEFT 0.4.0
|
|
|
|
| 211 |
|
| 212 |
- PEFT 0.4.0
|
|
|
|
| 180 |
- bnb_4bit_use_double_quant: True
|
| 181 |
- bnb_4bit_compute_dtype: bfloat16
|
| 182 |
|
| 183 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 184 |
+
- load_in_8bit: False
|
| 185 |
+
- load_in_4bit: True
|
| 186 |
+
- llm_int8_threshold: 6.0
|
| 187 |
+
- llm_int8_skip_modules: None
|
| 188 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 189 |
+
- llm_int8_has_fp16_weight: False
|
| 190 |
+
- bnb_4bit_quant_type: nf4
|
| 191 |
+
- bnb_4bit_use_double_quant: True
|
| 192 |
+
- bnb_4bit_compute_dtype: bfloat16
|
| 193 |
+
|
| 194 |
The following `bitsandbytes` quantization config was used during training:
|
| 195 |
- load_in_8bit: False
|
| 196 |
- load_in_4bit: True
|
|
|
|
| 219 |
- PEFT 0.4.0
|
| 220 |
- PEFT 0.4.0
|
| 221 |
- PEFT 0.4.0
|
| 222 |
+
- PEFT 0.4.0
|
| 223 |
|
| 224 |
- PEFT 0.4.0
|
checkpoint-6200/adapter_model/adapter_model/adapter_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 319977229
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e89c2ce43f512a93774b45f193b0dd939c1fb7cb618f77f34bb9707caf0f4b39
|
| 3 |
size 319977229
|
{checkpoint-7800 → checkpoint-9800}/README.md
RENAMED
|
File without changes
|
{checkpoint-7800 → checkpoint-9800}/adapter_config.json
RENAMED
|
File without changes
|
{checkpoint-7800 → checkpoint-9800}/adapter_model.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 319977229
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:001215d79dce3fe3434e00202a8d7189e8f6c0796c7f37e48480e76aa0ae11a0
|
| 3 |
size 319977229
|
{checkpoint-7800 → checkpoint-9800}/added_tokens.json
RENAMED
|
File without changes
|
{checkpoint-7800 → checkpoint-9800}/optimizer.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1279539973
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aaecc676d94724406afb25421eac67634fb15bb987212cb8eccc48448e6d2c58
|
| 3 |
size 1279539973
|
{checkpoint-7800 → checkpoint-9800}/rng_state.pth
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14511
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:851f72ee80aa89583a69bfc2ece2a5aac4925674df7f29a4b85d2ed8011419d5
|
| 3 |
size 14511
|
{checkpoint-7800 → checkpoint-9800}/scheduler.pt
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 627
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0dca521772e835c5c16030c5b4c0d3720f6592faeca96d824ba30cf9f318f744
|
| 3 |
size 627
|
{checkpoint-7800 → checkpoint-9800}/special_tokens_map.json
RENAMED
|
File without changes
|
{checkpoint-7800 → checkpoint-9800}/tokenizer.model
RENAMED
|
File without changes
|
{checkpoint-7800 → checkpoint-9800}/tokenizer_config.json
RENAMED
|
File without changes
|
{checkpoint-7800 → checkpoint-9800}/trainer_state.json
RENAMED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": 0.7293602228164673,
|
| 3 |
"best_model_checkpoint": "experts/expert-16/checkpoint-6200",
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
@@ -7455,11 +7455,1921 @@
|
|
| 7455 |
"mmlu_eval_accuracy_world_religions": 0.6842105263157895,
|
| 7456 |
"mmlu_loss": 1.4828916400204128,
|
| 7457 |
"step": 7800
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7458 |
}
|
| 7459 |
],
|
| 7460 |
"max_steps": 10000,
|
| 7461 |
"num_train_epochs": 4,
|
| 7462 |
-
"total_flos": 2.
|
| 7463 |
"trial_name": null,
|
| 7464 |
"trial_params": null
|
| 7465 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": 0.7293602228164673,
|
| 3 |
"best_model_checkpoint": "experts/expert-16/checkpoint-6200",
|
| 4 |
+
"epoch": 3.1051964512040557,
|
| 5 |
+
"global_step": 9800,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
|
|
|
| 7455 |
"mmlu_eval_accuracy_world_religions": 0.6842105263157895,
|
| 7456 |
"mmlu_loss": 1.4828916400204128,
|
| 7457 |
"step": 7800
|
| 7458 |
+
},
|
| 7459 |
+
{
|
| 7460 |
+
"epoch": 2.47,
|
| 7461 |
+
"learning_rate": 0.0002,
|
| 7462 |
+
"loss": 0.6005,
|
| 7463 |
+
"step": 7810
|
| 7464 |
+
},
|
| 7465 |
+
{
|
| 7466 |
+
"epoch": 2.48,
|
| 7467 |
+
"learning_rate": 0.0002,
|
| 7468 |
+
"loss": 0.6662,
|
| 7469 |
+
"step": 7820
|
| 7470 |
+
},
|
| 7471 |
+
{
|
| 7472 |
+
"epoch": 2.48,
|
| 7473 |
+
"learning_rate": 0.0002,
|
| 7474 |
+
"loss": 0.5821,
|
| 7475 |
+
"step": 7830
|
| 7476 |
+
},
|
| 7477 |
+
{
|
| 7478 |
+
"epoch": 2.48,
|
| 7479 |
+
"learning_rate": 0.0002,
|
| 7480 |
+
"loss": 0.5826,
|
| 7481 |
+
"step": 7840
|
| 7482 |
+
},
|
| 7483 |
+
{
|
| 7484 |
+
"epoch": 2.49,
|
| 7485 |
+
"learning_rate": 0.0002,
|
| 7486 |
+
"loss": 0.5804,
|
| 7487 |
+
"step": 7850
|
| 7488 |
+
},
|
| 7489 |
+
{
|
| 7490 |
+
"epoch": 2.49,
|
| 7491 |
+
"learning_rate": 0.0002,
|
| 7492 |
+
"loss": 0.587,
|
| 7493 |
+
"step": 7860
|
| 7494 |
+
},
|
| 7495 |
+
{
|
| 7496 |
+
"epoch": 2.49,
|
| 7497 |
+
"learning_rate": 0.0002,
|
| 7498 |
+
"loss": 0.6062,
|
| 7499 |
+
"step": 7870
|
| 7500 |
+
},
|
| 7501 |
+
{
|
| 7502 |
+
"epoch": 2.5,
|
| 7503 |
+
"learning_rate": 0.0002,
|
| 7504 |
+
"loss": 0.5616,
|
| 7505 |
+
"step": 7880
|
| 7506 |
+
},
|
| 7507 |
+
{
|
| 7508 |
+
"epoch": 2.5,
|
| 7509 |
+
"learning_rate": 0.0002,
|
| 7510 |
+
"loss": 0.6351,
|
| 7511 |
+
"step": 7890
|
| 7512 |
+
},
|
| 7513 |
+
{
|
| 7514 |
+
"epoch": 2.5,
|
| 7515 |
+
"learning_rate": 0.0002,
|
| 7516 |
+
"loss": 0.5738,
|
| 7517 |
+
"step": 7900
|
| 7518 |
+
},
|
| 7519 |
+
{
|
| 7520 |
+
"epoch": 2.51,
|
| 7521 |
+
"learning_rate": 0.0002,
|
| 7522 |
+
"loss": 0.5564,
|
| 7523 |
+
"step": 7910
|
| 7524 |
+
},
|
| 7525 |
+
{
|
| 7526 |
+
"epoch": 2.51,
|
| 7527 |
+
"learning_rate": 0.0002,
|
| 7528 |
+
"loss": 0.5696,
|
| 7529 |
+
"step": 7920
|
| 7530 |
+
},
|
| 7531 |
+
{
|
| 7532 |
+
"epoch": 2.51,
|
| 7533 |
+
"learning_rate": 0.0002,
|
| 7534 |
+
"loss": 0.5812,
|
| 7535 |
+
"step": 7930
|
| 7536 |
+
},
|
| 7537 |
+
{
|
| 7538 |
+
"epoch": 2.52,
|
| 7539 |
+
"learning_rate": 0.0002,
|
| 7540 |
+
"loss": 0.5786,
|
| 7541 |
+
"step": 7940
|
| 7542 |
+
},
|
| 7543 |
+
{
|
| 7544 |
+
"epoch": 2.52,
|
| 7545 |
+
"learning_rate": 0.0002,
|
| 7546 |
+
"loss": 0.6053,
|
| 7547 |
+
"step": 7950
|
| 7548 |
+
},
|
| 7549 |
+
{
|
| 7550 |
+
"epoch": 2.52,
|
| 7551 |
+
"learning_rate": 0.0002,
|
| 7552 |
+
"loss": 0.5727,
|
| 7553 |
+
"step": 7960
|
| 7554 |
+
},
|
| 7555 |
+
{
|
| 7556 |
+
"epoch": 2.53,
|
| 7557 |
+
"learning_rate": 0.0002,
|
| 7558 |
+
"loss": 0.621,
|
| 7559 |
+
"step": 7970
|
| 7560 |
+
},
|
| 7561 |
+
{
|
| 7562 |
+
"epoch": 2.53,
|
| 7563 |
+
"learning_rate": 0.0002,
|
| 7564 |
+
"loss": 0.5679,
|
| 7565 |
+
"step": 7980
|
| 7566 |
+
},
|
| 7567 |
+
{
|
| 7568 |
+
"epoch": 2.53,
|
| 7569 |
+
"learning_rate": 0.0002,
|
| 7570 |
+
"loss": 0.6138,
|
| 7571 |
+
"step": 7990
|
| 7572 |
+
},
|
| 7573 |
+
{
|
| 7574 |
+
"epoch": 2.53,
|
| 7575 |
+
"learning_rate": 0.0002,
|
| 7576 |
+
"loss": 0.588,
|
| 7577 |
+
"step": 8000
|
| 7578 |
+
},
|
| 7579 |
+
{
|
| 7580 |
+
"epoch": 2.53,
|
| 7581 |
+
"eval_loss": 0.7585816979408264,
|
| 7582 |
+
"eval_runtime": 111.2835,
|
| 7583 |
+
"eval_samples_per_second": 8.986,
|
| 7584 |
+
"eval_steps_per_second": 4.493,
|
| 7585 |
+
"step": 8000
|
| 7586 |
+
},
|
| 7587 |
+
{
|
| 7588 |
+
"epoch": 2.53,
|
| 7589 |
+
"mmlu_eval_accuracy": 0.48589851563960756,
|
| 7590 |
+
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
|
| 7591 |
+
"mmlu_eval_accuracy_anatomy": 0.6428571428571429,
|
| 7592 |
+
"mmlu_eval_accuracy_astronomy": 0.4375,
|
| 7593 |
+
"mmlu_eval_accuracy_business_ethics": 0.45454545454545453,
|
| 7594 |
+
"mmlu_eval_accuracy_clinical_knowledge": 0.5172413793103449,
|
| 7595 |
+
"mmlu_eval_accuracy_college_biology": 0.375,
|
| 7596 |
+
"mmlu_eval_accuracy_college_chemistry": 0.375,
|
| 7597 |
+
"mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
|
| 7598 |
+
"mmlu_eval_accuracy_college_mathematics": 0.09090909090909091,
|
| 7599 |
+
"mmlu_eval_accuracy_college_medicine": 0.4090909090909091,
|
| 7600 |
+
"mmlu_eval_accuracy_college_physics": 0.18181818181818182,
|
| 7601 |
+
"mmlu_eval_accuracy_computer_security": 0.45454545454545453,
|
| 7602 |
+
"mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
|
| 7603 |
+
"mmlu_eval_accuracy_econometrics": 0.25,
|
| 7604 |
+
"mmlu_eval_accuracy_electrical_engineering": 0.375,
|
| 7605 |
+
"mmlu_eval_accuracy_elementary_mathematics": 0.34146341463414637,
|
| 7606 |
+
"mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
|
| 7607 |
+
"mmlu_eval_accuracy_global_facts": 0.5,
|
| 7608 |
+
"mmlu_eval_accuracy_high_school_biology": 0.40625,
|
| 7609 |
+
"mmlu_eval_accuracy_high_school_chemistry": 0.22727272727272727,
|
| 7610 |
+
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
|
| 7611 |
+
"mmlu_eval_accuracy_high_school_european_history": 0.6111111111111112,
|
| 7612 |
+
"mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
|
| 7613 |
+
"mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
|
| 7614 |
+
"mmlu_eval_accuracy_high_school_macroeconomics": 0.4883720930232558,
|
| 7615 |
+
"mmlu_eval_accuracy_high_school_mathematics": 0.1724137931034483,
|
| 7616 |
+
"mmlu_eval_accuracy_high_school_microeconomics": 0.5384615384615384,
|
| 7617 |
+
"mmlu_eval_accuracy_high_school_physics": 0.0,
|
| 7618 |
+
"mmlu_eval_accuracy_high_school_psychology": 0.9,
|
| 7619 |
+
"mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
|
| 7620 |
+
"mmlu_eval_accuracy_high_school_us_history": 0.6363636363636364,
|
| 7621 |
+
"mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
|
| 7622 |
+
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
|
| 7623 |
+
"mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
|
| 7624 |
+
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
|
| 7625 |
+
"mmlu_eval_accuracy_jurisprudence": 0.36363636363636365,
|
| 7626 |
+
"mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
|
| 7627 |
+
"mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
|
| 7628 |
+
"mmlu_eval_accuracy_management": 0.6363636363636364,
|
| 7629 |
+
"mmlu_eval_accuracy_marketing": 0.68,
|
| 7630 |
+
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
|
| 7631 |
+
"mmlu_eval_accuracy_miscellaneous": 0.6627906976744186,
|
| 7632 |
+
"mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
|
| 7633 |
+
"mmlu_eval_accuracy_moral_scenarios": 0.25,
|
| 7634 |
+
"mmlu_eval_accuracy_nutrition": 0.48484848484848486,
|
| 7635 |
+
"mmlu_eval_accuracy_philosophy": 0.5,
|
| 7636 |
+
"mmlu_eval_accuracy_prehistory": 0.5428571428571428,
|
| 7637 |
+
"mmlu_eval_accuracy_professional_accounting": 0.3225806451612903,
|
| 7638 |
+
"mmlu_eval_accuracy_professional_law": 0.3588235294117647,
|
| 7639 |
+
"mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
|
| 7640 |
+
"mmlu_eval_accuracy_professional_psychology": 0.4927536231884058,
|
| 7641 |
+
"mmlu_eval_accuracy_public_relations": 0.5,
|
| 7642 |
+
"mmlu_eval_accuracy_security_studies": 0.4074074074074074,
|
| 7643 |
+
"mmlu_eval_accuracy_sociology": 0.7272727272727273,
|
| 7644 |
+
"mmlu_eval_accuracy_us_foreign_policy": 0.6363636363636364,
|
| 7645 |
+
"mmlu_eval_accuracy_virology": 0.5,
|
| 7646 |
+
"mmlu_eval_accuracy_world_religions": 0.631578947368421,
|
| 7647 |
+
"mmlu_loss": 1.566373301120402,
|
| 7648 |
+
"step": 8000
|
| 7649 |
+
},
|
| 7650 |
+
{
|
| 7651 |
+
"epoch": 2.54,
|
| 7652 |
+
"learning_rate": 0.0002,
|
| 7653 |
+
"loss": 0.5624,
|
| 7654 |
+
"step": 8010
|
| 7655 |
+
},
|
| 7656 |
+
{
|
| 7657 |
+
"epoch": 2.54,
|
| 7658 |
+
"learning_rate": 0.0002,
|
| 7659 |
+
"loss": 0.6206,
|
| 7660 |
+
"step": 8020
|
| 7661 |
+
},
|
| 7662 |
+
{
|
| 7663 |
+
"epoch": 2.54,
|
| 7664 |
+
"learning_rate": 0.0002,
|
| 7665 |
+
"loss": 0.607,
|
| 7666 |
+
"step": 8030
|
| 7667 |
+
},
|
| 7668 |
+
{
|
| 7669 |
+
"epoch": 2.55,
|
| 7670 |
+
"learning_rate": 0.0002,
|
| 7671 |
+
"loss": 0.6344,
|
| 7672 |
+
"step": 8040
|
| 7673 |
+
},
|
| 7674 |
+
{
|
| 7675 |
+
"epoch": 2.55,
|
| 7676 |
+
"learning_rate": 0.0002,
|
| 7677 |
+
"loss": 0.6705,
|
| 7678 |
+
"step": 8050
|
| 7679 |
+
},
|
| 7680 |
+
{
|
| 7681 |
+
"epoch": 2.55,
|
| 7682 |
+
"learning_rate": 0.0002,
|
| 7683 |
+
"loss": 0.5679,
|
| 7684 |
+
"step": 8060
|
| 7685 |
+
},
|
| 7686 |
+
{
|
| 7687 |
+
"epoch": 2.56,
|
| 7688 |
+
"learning_rate": 0.0002,
|
| 7689 |
+
"loss": 0.6,
|
| 7690 |
+
"step": 8070
|
| 7691 |
+
},
|
| 7692 |
+
{
|
| 7693 |
+
"epoch": 2.56,
|
| 7694 |
+
"learning_rate": 0.0002,
|
| 7695 |
+
"loss": 0.6486,
|
| 7696 |
+
"step": 8080
|
| 7697 |
+
},
|
| 7698 |
+
{
|
| 7699 |
+
"epoch": 2.56,
|
| 7700 |
+
"learning_rate": 0.0002,
|
| 7701 |
+
"loss": 0.5959,
|
| 7702 |
+
"step": 8090
|
| 7703 |
+
},
|
| 7704 |
+
{
|
| 7705 |
+
"epoch": 2.57,
|
| 7706 |
+
"learning_rate": 0.0002,
|
| 7707 |
+
"loss": 0.6454,
|
| 7708 |
+
"step": 8100
|
| 7709 |
+
},
|
| 7710 |
+
{
|
| 7711 |
+
"epoch": 2.57,
|
| 7712 |
+
"learning_rate": 0.0002,
|
| 7713 |
+
"loss": 0.6085,
|
| 7714 |
+
"step": 8110
|
| 7715 |
+
},
|
| 7716 |
+
{
|
| 7717 |
+
"epoch": 2.57,
|
| 7718 |
+
"learning_rate": 0.0002,
|
| 7719 |
+
"loss": 0.5509,
|
| 7720 |
+
"step": 8120
|
| 7721 |
+
},
|
| 7722 |
+
{
|
| 7723 |
+
"epoch": 2.58,
|
| 7724 |
+
"learning_rate": 0.0002,
|
| 7725 |
+
"loss": 0.6267,
|
| 7726 |
+
"step": 8130
|
| 7727 |
+
},
|
| 7728 |
+
{
|
| 7729 |
+
"epoch": 2.58,
|
| 7730 |
+
"learning_rate": 0.0002,
|
| 7731 |
+
"loss": 0.5865,
|
| 7732 |
+
"step": 8140
|
| 7733 |
+
},
|
| 7734 |
+
{
|
| 7735 |
+
"epoch": 2.58,
|
| 7736 |
+
"learning_rate": 0.0002,
|
| 7737 |
+
"loss": 0.6002,
|
| 7738 |
+
"step": 8150
|
| 7739 |
+
},
|
| 7740 |
+
{
|
| 7741 |
+
"epoch": 2.59,
|
| 7742 |
+
"learning_rate": 0.0002,
|
| 7743 |
+
"loss": 0.6342,
|
| 7744 |
+
"step": 8160
|
| 7745 |
+
},
|
| 7746 |
+
{
|
| 7747 |
+
"epoch": 2.59,
|
| 7748 |
+
"learning_rate": 0.0002,
|
| 7749 |
+
"loss": 0.6312,
|
| 7750 |
+
"step": 8170
|
| 7751 |
+
},
|
| 7752 |
+
{
|
| 7753 |
+
"epoch": 2.59,
|
| 7754 |
+
"learning_rate": 0.0002,
|
| 7755 |
+
"loss": 0.6361,
|
| 7756 |
+
"step": 8180
|
| 7757 |
+
},
|
| 7758 |
+
{
|
| 7759 |
+
"epoch": 2.6,
|
| 7760 |
+
"learning_rate": 0.0002,
|
| 7761 |
+
"loss": 0.5676,
|
| 7762 |
+
"step": 8190
|
| 7763 |
+
},
|
| 7764 |
+
{
|
| 7765 |
+
"epoch": 2.6,
|
| 7766 |
+
"learning_rate": 0.0002,
|
| 7767 |
+
"loss": 0.6125,
|
| 7768 |
+
"step": 8200
|
| 7769 |
+
},
|
| 7770 |
+
{
|
| 7771 |
+
"epoch": 2.6,
|
| 7772 |
+
"eval_loss": 0.7568719387054443,
|
| 7773 |
+
"eval_runtime": 111.2374,
|
| 7774 |
+
"eval_samples_per_second": 8.99,
|
| 7775 |
+
"eval_steps_per_second": 4.495,
|
| 7776 |
+
"step": 8200
|
| 7777 |
+
},
|
| 7778 |
+
{
|
| 7779 |
+
"epoch": 2.6,
|
| 7780 |
+
"mmlu_eval_accuracy": 0.4699982014237092,
|
| 7781 |
+
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
|
| 7782 |
+
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
|
| 7783 |
+
"mmlu_eval_accuracy_astronomy": 0.4375,
|
| 7784 |
+
"mmlu_eval_accuracy_business_ethics": 0.45454545454545453,
|
| 7785 |
+
"mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
|
| 7786 |
+
"mmlu_eval_accuracy_college_biology": 0.375,
|
| 7787 |
+
"mmlu_eval_accuracy_college_chemistry": 0.375,
|
| 7788 |
+
"mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
|
| 7789 |
+
"mmlu_eval_accuracy_college_mathematics": 0.09090909090909091,
|
| 7790 |
+
"mmlu_eval_accuracy_college_medicine": 0.4090909090909091,
|
| 7791 |
+
"mmlu_eval_accuracy_college_physics": 0.2727272727272727,
|
| 7792 |
+
"mmlu_eval_accuracy_computer_security": 0.36363636363636365,
|
| 7793 |
+
"mmlu_eval_accuracy_conceptual_physics": 0.38461538461538464,
|
| 7794 |
+
"mmlu_eval_accuracy_econometrics": 0.25,
|
| 7795 |
+
"mmlu_eval_accuracy_electrical_engineering": 0.25,
|
| 7796 |
+
"mmlu_eval_accuracy_elementary_mathematics": 0.34146341463414637,
|
| 7797 |
+
"mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
|
| 7798 |
+
"mmlu_eval_accuracy_global_facts": 0.3,
|
| 7799 |
+
"mmlu_eval_accuracy_high_school_biology": 0.375,
|
| 7800 |
+
"mmlu_eval_accuracy_high_school_chemistry": 0.2727272727272727,
|
| 7801 |
+
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
|
| 7802 |
+
"mmlu_eval_accuracy_high_school_european_history": 0.5555555555555556,
|
| 7803 |
+
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
|
| 7804 |
+
"mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
|
| 7805 |
+
"mmlu_eval_accuracy_high_school_macroeconomics": 0.46511627906976744,
|
| 7806 |
+
"mmlu_eval_accuracy_high_school_mathematics": 0.20689655172413793,
|
| 7807 |
+
"mmlu_eval_accuracy_high_school_microeconomics": 0.46153846153846156,
|
| 7808 |
+
"mmlu_eval_accuracy_high_school_physics": 0.058823529411764705,
|
| 7809 |
+
"mmlu_eval_accuracy_high_school_psychology": 0.8666666666666667,
|
| 7810 |
+
"mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
|
| 7811 |
+
"mmlu_eval_accuracy_high_school_us_history": 0.6363636363636364,
|
| 7812 |
+
"mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
|
| 7813 |
+
"mmlu_eval_accuracy_human_aging": 0.6956521739130435,
|
| 7814 |
+
"mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
|
| 7815 |
+
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
|
| 7816 |
+
"mmlu_eval_accuracy_jurisprudence": 0.36363636363636365,
|
| 7817 |
+
"mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
|
| 7818 |
+
"mmlu_eval_accuracy_machine_learning": 0.18181818181818182,
|
| 7819 |
+
"mmlu_eval_accuracy_management": 0.6363636363636364,
|
| 7820 |
+
"mmlu_eval_accuracy_marketing": 0.68,
|
| 7821 |
+
"mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
|
| 7822 |
+
"mmlu_eval_accuracy_miscellaneous": 0.6744186046511628,
|
| 7823 |
+
"mmlu_eval_accuracy_moral_disputes": 0.47368421052631576,
|
| 7824 |
+
"mmlu_eval_accuracy_moral_scenarios": 0.24,
|
| 7825 |
+
"mmlu_eval_accuracy_nutrition": 0.6060606060606061,
|
| 7826 |
+
"mmlu_eval_accuracy_philosophy": 0.47058823529411764,
|
| 7827 |
+
"mmlu_eval_accuracy_prehistory": 0.5142857142857142,
|
| 7828 |
+
"mmlu_eval_accuracy_professional_accounting": 0.2903225806451613,
|
| 7829 |
+
"mmlu_eval_accuracy_professional_law": 0.3588235294117647,
|
| 7830 |
+
"mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
|
| 7831 |
+
"mmlu_eval_accuracy_professional_psychology": 0.5072463768115942,
|
| 7832 |
+
"mmlu_eval_accuracy_public_relations": 0.5833333333333334,
|
| 7833 |
+
"mmlu_eval_accuracy_security_studies": 0.4444444444444444,
|
| 7834 |
+
"mmlu_eval_accuracy_sociology": 0.6818181818181818,
|
| 7835 |
+
"mmlu_eval_accuracy_us_foreign_policy": 0.6363636363636364,
|
| 7836 |
+
"mmlu_eval_accuracy_virology": 0.5,
|
| 7837 |
+
"mmlu_eval_accuracy_world_religions": 0.6842105263157895,
|
| 7838 |
+
"mmlu_loss": 1.3685555242527248,
|
| 7839 |
+
"step": 8200
|
| 7840 |
+
},
|
| 7841 |
+
{
|
| 7842 |
+
"epoch": 2.6,
|
| 7843 |
+
"learning_rate": 0.0002,
|
| 7844 |
+
"loss": 0.5992,
|
| 7845 |
+
"step": 8210
|
| 7846 |
+
},
|
| 7847 |
+
{
|
| 7848 |
+
"epoch": 2.6,
|
| 7849 |
+
"learning_rate": 0.0002,
|
| 7850 |
+
"loss": 0.6068,
|
| 7851 |
+
"step": 8220
|
| 7852 |
+
},
|
| 7853 |
+
{
|
| 7854 |
+
"epoch": 2.61,
|
| 7855 |
+
"learning_rate": 0.0002,
|
| 7856 |
+
"loss": 0.6986,
|
| 7857 |
+
"step": 8230
|
| 7858 |
+
},
|
| 7859 |
+
{
|
| 7860 |
+
"epoch": 2.61,
|
| 7861 |
+
"learning_rate": 0.0002,
|
| 7862 |
+
"loss": 0.5809,
|
| 7863 |
+
"step": 8240
|
| 7864 |
+
},
|
| 7865 |
+
{
|
| 7866 |
+
"epoch": 2.61,
|
| 7867 |
+
"learning_rate": 0.0002,
|
| 7868 |
+
"loss": 0.6368,
|
| 7869 |
+
"step": 8250
|
| 7870 |
+
},
|
| 7871 |
+
{
|
| 7872 |
+
"epoch": 2.62,
|
| 7873 |
+
"learning_rate": 0.0002,
|
| 7874 |
+
"loss": 0.5731,
|
| 7875 |
+
"step": 8260
|
| 7876 |
+
},
|
| 7877 |
+
{
|
| 7878 |
+
"epoch": 2.62,
|
| 7879 |
+
"learning_rate": 0.0002,
|
| 7880 |
+
"loss": 0.6439,
|
| 7881 |
+
"step": 8270
|
| 7882 |
+
},
|
| 7883 |
+
{
|
| 7884 |
+
"epoch": 2.62,
|
| 7885 |
+
"learning_rate": 0.0002,
|
| 7886 |
+
"loss": 0.5661,
|
| 7887 |
+
"step": 8280
|
| 7888 |
+
},
|
| 7889 |
+
{
|
| 7890 |
+
"epoch": 2.63,
|
| 7891 |
+
"learning_rate": 0.0002,
|
| 7892 |
+
"loss": 0.5816,
|
| 7893 |
+
"step": 8290
|
| 7894 |
+
},
|
| 7895 |
+
{
|
| 7896 |
+
"epoch": 2.63,
|
| 7897 |
+
"learning_rate": 0.0002,
|
| 7898 |
+
"loss": 0.5385,
|
| 7899 |
+
"step": 8300
|
| 7900 |
+
},
|
| 7901 |
+
{
|
| 7902 |
+
"epoch": 2.63,
|
| 7903 |
+
"learning_rate": 0.0002,
|
| 7904 |
+
"loss": 0.5913,
|
| 7905 |
+
"step": 8310
|
| 7906 |
+
},
|
| 7907 |
+
{
|
| 7908 |
+
"epoch": 2.64,
|
| 7909 |
+
"learning_rate": 0.0002,
|
| 7910 |
+
"loss": 0.5817,
|
| 7911 |
+
"step": 8320
|
| 7912 |
+
},
|
| 7913 |
+
{
|
| 7914 |
+
"epoch": 2.64,
|
| 7915 |
+
"learning_rate": 0.0002,
|
| 7916 |
+
"loss": 0.6098,
|
| 7917 |
+
"step": 8330
|
| 7918 |
+
},
|
| 7919 |
+
{
|
| 7920 |
+
"epoch": 2.64,
|
| 7921 |
+
"learning_rate": 0.0002,
|
| 7922 |
+
"loss": 0.558,
|
| 7923 |
+
"step": 8340
|
| 7924 |
+
},
|
| 7925 |
+
{
|
| 7926 |
+
"epoch": 2.65,
|
| 7927 |
+
"learning_rate": 0.0002,
|
| 7928 |
+
"loss": 0.6008,
|
| 7929 |
+
"step": 8350
|
| 7930 |
+
},
|
| 7931 |
+
{
|
| 7932 |
+
"epoch": 2.65,
|
| 7933 |
+
"learning_rate": 0.0002,
|
| 7934 |
+
"loss": 0.5921,
|
| 7935 |
+
"step": 8360
|
| 7936 |
+
},
|
| 7937 |
+
{
|
| 7938 |
+
"epoch": 2.65,
|
| 7939 |
+
"learning_rate": 0.0002,
|
| 7940 |
+
"loss": 0.6194,
|
| 7941 |
+
"step": 8370
|
| 7942 |
+
},
|
| 7943 |
+
{
|
| 7944 |
+
"epoch": 2.66,
|
| 7945 |
+
"learning_rate": 0.0002,
|
| 7946 |
+
"loss": 0.6849,
|
| 7947 |
+
"step": 8380
|
| 7948 |
+
},
|
| 7949 |
+
{
|
| 7950 |
+
"epoch": 2.66,
|
| 7951 |
+
"learning_rate": 0.0002,
|
| 7952 |
+
"loss": 0.5851,
|
| 7953 |
+
"step": 8390
|
| 7954 |
+
},
|
| 7955 |
+
{
|
| 7956 |
+
"epoch": 2.66,
|
| 7957 |
+
"learning_rate": 0.0002,
|
| 7958 |
+
"loss": 0.5574,
|
| 7959 |
+
"step": 8400
|
| 7960 |
+
},
|
| 7961 |
+
{
|
| 7962 |
+
"epoch": 2.66,
|
| 7963 |
+
"eval_loss": 0.7574586868286133,
|
| 7964 |
+
"eval_runtime": 111.0853,
|
| 7965 |
+
"eval_samples_per_second": 9.002,
|
| 7966 |
+
"eval_steps_per_second": 4.501,
|
| 7967 |
+
"step": 8400
|
| 7968 |
+
},
|
| 7969 |
+
{
|
| 7970 |
+
"epoch": 2.66,
|
| 7971 |
+
"mmlu_eval_accuracy": 0.47813110611906134,
|
| 7972 |
+
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
|
| 7973 |
+
"mmlu_eval_accuracy_anatomy": 0.5,
|
| 7974 |
+
"mmlu_eval_accuracy_astronomy": 0.4375,
|
| 7975 |
+
"mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
|
| 7976 |
+
"mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
|
| 7977 |
+
"mmlu_eval_accuracy_college_biology": 0.375,
|
| 7978 |
+
"mmlu_eval_accuracy_college_chemistry": 0.375,
|
| 7979 |
+
"mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
|
| 7980 |
+
"mmlu_eval_accuracy_college_mathematics": 0.09090909090909091,
|
| 7981 |
+
"mmlu_eval_accuracy_college_medicine": 0.4090909090909091,
|
| 7982 |
+
"mmlu_eval_accuracy_college_physics": 0.18181818181818182,
|
| 7983 |
+
"mmlu_eval_accuracy_computer_security": 0.45454545454545453,
|
| 7984 |
+
"mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
|
| 7985 |
+
"mmlu_eval_accuracy_econometrics": 0.3333333333333333,
|
| 7986 |
+
"mmlu_eval_accuracy_electrical_engineering": 0.25,
|
| 7987 |
+
"mmlu_eval_accuracy_elementary_mathematics": 0.3170731707317073,
|
| 7988 |
+
"mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
|
| 7989 |
+
"mmlu_eval_accuracy_global_facts": 0.3,
|
| 7990 |
+
"mmlu_eval_accuracy_high_school_biology": 0.375,
|
| 7991 |
+
"mmlu_eval_accuracy_high_school_chemistry": 0.2727272727272727,
|
| 7992 |
+
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
|
| 7993 |
+
"mmlu_eval_accuracy_high_school_european_history": 0.6111111111111112,
|
| 7994 |
+
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
|
| 7995 |
+
"mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
|
| 7996 |
+
"mmlu_eval_accuracy_high_school_macroeconomics": 0.4883720930232558,
|
| 7997 |
+
"mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
|
| 7998 |
+
"mmlu_eval_accuracy_high_school_microeconomics": 0.5384615384615384,
|
| 7999 |
+
"mmlu_eval_accuracy_high_school_physics": 0.058823529411764705,
|
| 8000 |
+
"mmlu_eval_accuracy_high_school_psychology": 0.8833333333333333,
|
| 8001 |
+
"mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
|
| 8002 |
+
"mmlu_eval_accuracy_high_school_us_history": 0.6363636363636364,
|
| 8003 |
+
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
|
| 8004 |
+
"mmlu_eval_accuracy_human_aging": 0.6956521739130435,
|
| 8005 |
+
"mmlu_eval_accuracy_human_sexuality": 0.3333333333333333,
|
| 8006 |
+
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
|
| 8007 |
+
"mmlu_eval_accuracy_jurisprudence": 0.36363636363636365,
|
| 8008 |
+
"mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
|
| 8009 |
+
"mmlu_eval_accuracy_machine_learning": 0.18181818181818182,
|
| 8010 |
+
"mmlu_eval_accuracy_management": 0.6363636363636364,
|
| 8011 |
+
"mmlu_eval_accuracy_marketing": 0.76,
|
| 8012 |
+
"mmlu_eval_accuracy_medical_genetics": 0.8181818181818182,
|
| 8013 |
+
"mmlu_eval_accuracy_miscellaneous": 0.6395348837209303,
|
| 8014 |
+
"mmlu_eval_accuracy_moral_disputes": 0.47368421052631576,
|
| 8015 |
+
"mmlu_eval_accuracy_moral_scenarios": 0.25,
|
| 8016 |
+
"mmlu_eval_accuracy_nutrition": 0.6363636363636364,
|
| 8017 |
+
"mmlu_eval_accuracy_philosophy": 0.47058823529411764,
|
| 8018 |
+
"mmlu_eval_accuracy_prehistory": 0.5142857142857142,
|
| 8019 |
+
"mmlu_eval_accuracy_professional_accounting": 0.3225806451612903,
|
| 8020 |
+
"mmlu_eval_accuracy_professional_law": 0.3588235294117647,
|
| 8021 |
+
"mmlu_eval_accuracy_professional_medicine": 0.5483870967741935,
|
| 8022 |
+
"mmlu_eval_accuracy_professional_psychology": 0.5072463768115942,
|
| 8023 |
+
"mmlu_eval_accuracy_public_relations": 0.5833333333333334,
|
| 8024 |
+
"mmlu_eval_accuracy_security_studies": 0.4074074074074074,
|
| 8025 |
+
"mmlu_eval_accuracy_sociology": 0.6818181818181818,
|
| 8026 |
+
"mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
|
| 8027 |
+
"mmlu_eval_accuracy_virology": 0.5,
|
| 8028 |
+
"mmlu_eval_accuracy_world_religions": 0.6842105263157895,
|
| 8029 |
+
"mmlu_loss": 1.4071070384574622,
|
| 8030 |
+
"step": 8400
|
| 8031 |
+
},
|
| 8032 |
+
{
|
| 8033 |
+
"epoch": 2.66,
|
| 8034 |
+
"learning_rate": 0.0002,
|
| 8035 |
+
"loss": 0.5795,
|
| 8036 |
+
"step": 8410
|
| 8037 |
+
},
|
| 8038 |
+
{
|
| 8039 |
+
"epoch": 2.67,
|
| 8040 |
+
"learning_rate": 0.0002,
|
| 8041 |
+
"loss": 0.6109,
|
| 8042 |
+
"step": 8420
|
| 8043 |
+
},
|
| 8044 |
+
{
|
| 8045 |
+
"epoch": 2.67,
|
| 8046 |
+
"learning_rate": 0.0002,
|
| 8047 |
+
"loss": 0.6136,
|
| 8048 |
+
"step": 8430
|
| 8049 |
+
},
|
| 8050 |
+
{
|
| 8051 |
+
"epoch": 2.67,
|
| 8052 |
+
"learning_rate": 0.0002,
|
| 8053 |
+
"loss": 0.5795,
|
| 8054 |
+
"step": 8440
|
| 8055 |
+
},
|
| 8056 |
+
{
|
| 8057 |
+
"epoch": 2.68,
|
| 8058 |
+
"learning_rate": 0.0002,
|
| 8059 |
+
"loss": 0.5639,
|
| 8060 |
+
"step": 8450
|
| 8061 |
+
},
|
| 8062 |
+
{
|
| 8063 |
+
"epoch": 2.68,
|
| 8064 |
+
"learning_rate": 0.0002,
|
| 8065 |
+
"loss": 0.5869,
|
| 8066 |
+
"step": 8460
|
| 8067 |
+
},
|
| 8068 |
+
{
|
| 8069 |
+
"epoch": 2.68,
|
| 8070 |
+
"learning_rate": 0.0002,
|
| 8071 |
+
"loss": 0.5946,
|
| 8072 |
+
"step": 8470
|
| 8073 |
+
},
|
| 8074 |
+
{
|
| 8075 |
+
"epoch": 2.69,
|
| 8076 |
+
"learning_rate": 0.0002,
|
| 8077 |
+
"loss": 0.5745,
|
| 8078 |
+
"step": 8480
|
| 8079 |
+
},
|
| 8080 |
+
{
|
| 8081 |
+
"epoch": 2.69,
|
| 8082 |
+
"learning_rate": 0.0002,
|
| 8083 |
+
"loss": 0.573,
|
| 8084 |
+
"step": 8490
|
| 8085 |
+
},
|
| 8086 |
+
{
|
| 8087 |
+
"epoch": 2.69,
|
| 8088 |
+
"learning_rate": 0.0002,
|
| 8089 |
+
"loss": 0.5846,
|
| 8090 |
+
"step": 8500
|
| 8091 |
+
},
|
| 8092 |
+
{
|
| 8093 |
+
"epoch": 2.7,
|
| 8094 |
+
"learning_rate": 0.0002,
|
| 8095 |
+
"loss": 0.6058,
|
| 8096 |
+
"step": 8510
|
| 8097 |
+
},
|
| 8098 |
+
{
|
| 8099 |
+
"epoch": 2.7,
|
| 8100 |
+
"learning_rate": 0.0002,
|
| 8101 |
+
"loss": 0.5072,
|
| 8102 |
+
"step": 8520
|
| 8103 |
+
},
|
| 8104 |
+
{
|
| 8105 |
+
"epoch": 2.7,
|
| 8106 |
+
"learning_rate": 0.0002,
|
| 8107 |
+
"loss": 0.6296,
|
| 8108 |
+
"step": 8530
|
| 8109 |
+
},
|
| 8110 |
+
{
|
| 8111 |
+
"epoch": 2.71,
|
| 8112 |
+
"learning_rate": 0.0002,
|
| 8113 |
+
"loss": 0.6057,
|
| 8114 |
+
"step": 8540
|
| 8115 |
+
},
|
| 8116 |
+
{
|
| 8117 |
+
"epoch": 2.71,
|
| 8118 |
+
"learning_rate": 0.0002,
|
| 8119 |
+
"loss": 0.544,
|
| 8120 |
+
"step": 8550
|
| 8121 |
+
},
|
| 8122 |
+
{
|
| 8123 |
+
"epoch": 2.71,
|
| 8124 |
+
"learning_rate": 0.0002,
|
| 8125 |
+
"loss": 0.6256,
|
| 8126 |
+
"step": 8560
|
| 8127 |
+
},
|
| 8128 |
+
{
|
| 8129 |
+
"epoch": 2.72,
|
| 8130 |
+
"learning_rate": 0.0002,
|
| 8131 |
+
"loss": 0.6307,
|
| 8132 |
+
"step": 8570
|
| 8133 |
+
},
|
| 8134 |
+
{
|
| 8135 |
+
"epoch": 2.72,
|
| 8136 |
+
"learning_rate": 0.0002,
|
| 8137 |
+
"loss": 0.5717,
|
| 8138 |
+
"step": 8580
|
| 8139 |
+
},
|
| 8140 |
+
{
|
| 8141 |
+
"epoch": 2.72,
|
| 8142 |
+
"learning_rate": 0.0002,
|
| 8143 |
+
"loss": 0.5946,
|
| 8144 |
+
"step": 8590
|
| 8145 |
+
},
|
| 8146 |
+
{
|
| 8147 |
+
"epoch": 2.72,
|
| 8148 |
+
"learning_rate": 0.0002,
|
| 8149 |
+
"loss": 0.6025,
|
| 8150 |
+
"step": 8600
|
| 8151 |
+
},
|
| 8152 |
+
{
|
| 8153 |
+
"epoch": 2.72,
|
| 8154 |
+
"eval_loss": 0.7557567358016968,
|
| 8155 |
+
"eval_runtime": 111.5506,
|
| 8156 |
+
"eval_samples_per_second": 8.965,
|
| 8157 |
+
"eval_steps_per_second": 4.482,
|
| 8158 |
+
"step": 8600
|
| 8159 |
+
},
|
| 8160 |
+
{
|
| 8161 |
+
"epoch": 2.72,
|
| 8162 |
+
"mmlu_eval_accuracy": 0.4768331170577644,
|
| 8163 |
+
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
|
| 8164 |
+
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
|
| 8165 |
+
"mmlu_eval_accuracy_astronomy": 0.4375,
|
| 8166 |
+
"mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
|
| 8167 |
+
"mmlu_eval_accuracy_clinical_knowledge": 0.5862068965517241,
|
| 8168 |
+
"mmlu_eval_accuracy_college_biology": 0.375,
|
| 8169 |
+
"mmlu_eval_accuracy_college_chemistry": 0.375,
|
| 8170 |
+
"mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
|
| 8171 |
+
"mmlu_eval_accuracy_college_mathematics": 0.18181818181818182,
|
| 8172 |
+
"mmlu_eval_accuracy_college_medicine": 0.5,
|
| 8173 |
+
"mmlu_eval_accuracy_college_physics": 0.18181818181818182,
|
| 8174 |
+
"mmlu_eval_accuracy_computer_security": 0.45454545454545453,
|
| 8175 |
+
"mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
|
| 8176 |
+
"mmlu_eval_accuracy_econometrics": 0.3333333333333333,
|
| 8177 |
+
"mmlu_eval_accuracy_electrical_engineering": 0.25,
|
| 8178 |
+
"mmlu_eval_accuracy_elementary_mathematics": 0.34146341463414637,
|
| 8179 |
+
"mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
|
| 8180 |
+
"mmlu_eval_accuracy_global_facts": 0.3,
|
| 8181 |
+
"mmlu_eval_accuracy_high_school_biology": 0.40625,
|
| 8182 |
+
"mmlu_eval_accuracy_high_school_chemistry": 0.22727272727272727,
|
| 8183 |
+
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
|
| 8184 |
+
"mmlu_eval_accuracy_high_school_european_history": 0.6111111111111112,
|
| 8185 |
+
"mmlu_eval_accuracy_high_school_geography": 0.7727272727272727,
|
| 8186 |
+
"mmlu_eval_accuracy_high_school_government_and_politics": 0.6190476190476191,
|
| 8187 |
+
"mmlu_eval_accuracy_high_school_macroeconomics": 0.4883720930232558,
|
| 8188 |
+
"mmlu_eval_accuracy_high_school_mathematics": 0.1724137931034483,
|
| 8189 |
+
"mmlu_eval_accuracy_high_school_microeconomics": 0.5,
|
| 8190 |
+
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
|
| 8191 |
+
"mmlu_eval_accuracy_high_school_psychology": 0.9,
|
| 8192 |
+
"mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
|
| 8193 |
+
"mmlu_eval_accuracy_high_school_us_history": 0.6363636363636364,
|
| 8194 |
+
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
|
| 8195 |
+
"mmlu_eval_accuracy_human_aging": 0.6956521739130435,
|
| 8196 |
+
"mmlu_eval_accuracy_human_sexuality": 0.3333333333333333,
|
| 8197 |
+
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
|
| 8198 |
+
"mmlu_eval_accuracy_jurisprudence": 0.36363636363636365,
|
| 8199 |
+
"mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
|
| 8200 |
+
"mmlu_eval_accuracy_machine_learning": 0.09090909090909091,
|
| 8201 |
+
"mmlu_eval_accuracy_management": 0.6363636363636364,
|
| 8202 |
+
"mmlu_eval_accuracy_marketing": 0.76,
|
| 8203 |
+
"mmlu_eval_accuracy_medical_genetics": 0.8181818181818182,
|
| 8204 |
+
"mmlu_eval_accuracy_miscellaneous": 0.6395348837209303,
|
| 8205 |
+
"mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
|
| 8206 |
+
"mmlu_eval_accuracy_moral_scenarios": 0.23,
|
| 8207 |
+
"mmlu_eval_accuracy_nutrition": 0.6363636363636364,
|
| 8208 |
+
"mmlu_eval_accuracy_philosophy": 0.47058823529411764,
|
| 8209 |
+
"mmlu_eval_accuracy_prehistory": 0.5428571428571428,
|
| 8210 |
+
"mmlu_eval_accuracy_professional_accounting": 0.2903225806451613,
|
| 8211 |
+
"mmlu_eval_accuracy_professional_law": 0.3352941176470588,
|
| 8212 |
+
"mmlu_eval_accuracy_professional_medicine": 0.5483870967741935,
|
| 8213 |
+
"mmlu_eval_accuracy_professional_psychology": 0.4782608695652174,
|
| 8214 |
+
"mmlu_eval_accuracy_public_relations": 0.5833333333333334,
|
| 8215 |
+
"mmlu_eval_accuracy_security_studies": 0.4074074074074074,
|
| 8216 |
+
"mmlu_eval_accuracy_sociology": 0.6363636363636364,
|
| 8217 |
+
"mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
|
| 8218 |
+
"mmlu_eval_accuracy_virology": 0.5,
|
| 8219 |
+
"mmlu_eval_accuracy_world_religions": 0.6842105263157895,
|
| 8220 |
+
"mmlu_loss": 1.4071306704152036,
|
| 8221 |
+
"step": 8600
|
| 8222 |
+
},
|
| 8223 |
+
{
|
| 8224 |
+
"epoch": 2.73,
|
| 8225 |
+
"learning_rate": 0.0002,
|
| 8226 |
+
"loss": 0.6607,
|
| 8227 |
+
"step": 8610
|
| 8228 |
+
},
|
| 8229 |
+
{
|
| 8230 |
+
"epoch": 2.73,
|
| 8231 |
+
"learning_rate": 0.0002,
|
| 8232 |
+
"loss": 0.6087,
|
| 8233 |
+
"step": 8620
|
| 8234 |
+
},
|
| 8235 |
+
{
|
| 8236 |
+
"epoch": 2.73,
|
| 8237 |
+
"learning_rate": 0.0002,
|
| 8238 |
+
"loss": 0.6077,
|
| 8239 |
+
"step": 8630
|
| 8240 |
+
},
|
| 8241 |
+
{
|
| 8242 |
+
"epoch": 2.74,
|
| 8243 |
+
"learning_rate": 0.0002,
|
| 8244 |
+
"loss": 0.5746,
|
| 8245 |
+
"step": 8640
|
| 8246 |
+
},
|
| 8247 |
+
{
|
| 8248 |
+
"epoch": 2.74,
|
| 8249 |
+
"learning_rate": 0.0002,
|
| 8250 |
+
"loss": 0.6372,
|
| 8251 |
+
"step": 8650
|
| 8252 |
+
},
|
| 8253 |
+
{
|
| 8254 |
+
"epoch": 2.74,
|
| 8255 |
+
"learning_rate": 0.0002,
|
| 8256 |
+
"loss": 0.603,
|
| 8257 |
+
"step": 8660
|
| 8258 |
+
},
|
| 8259 |
+
{
|
| 8260 |
+
"epoch": 2.75,
|
| 8261 |
+
"learning_rate": 0.0002,
|
| 8262 |
+
"loss": 0.5913,
|
| 8263 |
+
"step": 8670
|
| 8264 |
+
},
|
| 8265 |
+
{
|
| 8266 |
+
"epoch": 2.75,
|
| 8267 |
+
"learning_rate": 0.0002,
|
| 8268 |
+
"loss": 0.664,
|
| 8269 |
+
"step": 8680
|
| 8270 |
+
},
|
| 8271 |
+
{
|
| 8272 |
+
"epoch": 2.75,
|
| 8273 |
+
"learning_rate": 0.0002,
|
| 8274 |
+
"loss": 0.5766,
|
| 8275 |
+
"step": 8690
|
| 8276 |
+
},
|
| 8277 |
+
{
|
| 8278 |
+
"epoch": 2.76,
|
| 8279 |
+
"learning_rate": 0.0002,
|
| 8280 |
+
"loss": 0.6316,
|
| 8281 |
+
"step": 8700
|
| 8282 |
+
},
|
| 8283 |
+
{
|
| 8284 |
+
"epoch": 2.76,
|
| 8285 |
+
"learning_rate": 0.0002,
|
| 8286 |
+
"loss": 0.5913,
|
| 8287 |
+
"step": 8710
|
| 8288 |
+
},
|
| 8289 |
+
{
|
| 8290 |
+
"epoch": 2.76,
|
| 8291 |
+
"learning_rate": 0.0002,
|
| 8292 |
+
"loss": 0.5974,
|
| 8293 |
+
"step": 8720
|
| 8294 |
+
},
|
| 8295 |
+
{
|
| 8296 |
+
"epoch": 2.77,
|
| 8297 |
+
"learning_rate": 0.0002,
|
| 8298 |
+
"loss": 0.6519,
|
| 8299 |
+
"step": 8730
|
| 8300 |
+
},
|
| 8301 |
+
{
|
| 8302 |
+
"epoch": 2.77,
|
| 8303 |
+
"learning_rate": 0.0002,
|
| 8304 |
+
"loss": 0.6502,
|
| 8305 |
+
"step": 8740
|
| 8306 |
+
},
|
| 8307 |
+
{
|
| 8308 |
+
"epoch": 2.77,
|
| 8309 |
+
"learning_rate": 0.0002,
|
| 8310 |
+
"loss": 0.6069,
|
| 8311 |
+
"step": 8750
|
| 8312 |
+
},
|
| 8313 |
+
{
|
| 8314 |
+
"epoch": 2.78,
|
| 8315 |
+
"learning_rate": 0.0002,
|
| 8316 |
+
"loss": 0.6073,
|
| 8317 |
+
"step": 8760
|
| 8318 |
+
},
|
| 8319 |
+
{
|
| 8320 |
+
"epoch": 2.78,
|
| 8321 |
+
"learning_rate": 0.0002,
|
| 8322 |
+
"loss": 0.5314,
|
| 8323 |
+
"step": 8770
|
| 8324 |
+
},
|
| 8325 |
+
{
|
| 8326 |
+
"epoch": 2.78,
|
| 8327 |
+
"learning_rate": 0.0002,
|
| 8328 |
+
"loss": 0.6515,
|
| 8329 |
+
"step": 8780
|
| 8330 |
+
},
|
| 8331 |
+
{
|
| 8332 |
+
"epoch": 2.79,
|
| 8333 |
+
"learning_rate": 0.0002,
|
| 8334 |
+
"loss": 0.6515,
|
| 8335 |
+
"step": 8790
|
| 8336 |
+
},
|
| 8337 |
+
{
|
| 8338 |
+
"epoch": 2.79,
|
| 8339 |
+
"learning_rate": 0.0002,
|
| 8340 |
+
"loss": 0.6188,
|
| 8341 |
+
"step": 8800
|
| 8342 |
+
},
|
| 8343 |
+
{
|
| 8344 |
+
"epoch": 2.79,
|
| 8345 |
+
"eval_loss": 0.7575909495353699,
|
| 8346 |
+
"eval_runtime": 111.2733,
|
| 8347 |
+
"eval_samples_per_second": 8.987,
|
| 8348 |
+
"eval_steps_per_second": 4.493,
|
| 8349 |
+
"step": 8800
|
| 8350 |
+
},
|
| 8351 |
+
{
|
| 8352 |
+
"epoch": 2.79,
|
| 8353 |
+
"mmlu_eval_accuracy": 0.489029874622582,
|
| 8354 |
+
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
|
| 8355 |
+
"mmlu_eval_accuracy_anatomy": 0.6428571428571429,
|
| 8356 |
+
"mmlu_eval_accuracy_astronomy": 0.375,
|
| 8357 |
+
"mmlu_eval_accuracy_business_ethics": 0.45454545454545453,
|
| 8358 |
+
"mmlu_eval_accuracy_clinical_knowledge": 0.6206896551724138,
|
| 8359 |
+
"mmlu_eval_accuracy_college_biology": 0.375,
|
| 8360 |
+
"mmlu_eval_accuracy_college_chemistry": 0.375,
|
| 8361 |
+
"mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
|
| 8362 |
+
"mmlu_eval_accuracy_college_mathematics": 0.18181818181818182,
|
| 8363 |
+
"mmlu_eval_accuracy_college_medicine": 0.4090909090909091,
|
| 8364 |
+
"mmlu_eval_accuracy_college_physics": 0.2727272727272727,
|
| 8365 |
+
"mmlu_eval_accuracy_computer_security": 0.5454545454545454,
|
| 8366 |
+
"mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
|
| 8367 |
+
"mmlu_eval_accuracy_econometrics": 0.25,
|
| 8368 |
+
"mmlu_eval_accuracy_electrical_engineering": 0.25,
|
| 8369 |
+
"mmlu_eval_accuracy_elementary_mathematics": 0.3170731707317073,
|
| 8370 |
+
"mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
|
| 8371 |
+
"mmlu_eval_accuracy_global_facts": 0.5,
|
| 8372 |
+
"mmlu_eval_accuracy_high_school_biology": 0.40625,
|
| 8373 |
+
"mmlu_eval_accuracy_high_school_chemistry": 0.22727272727272727,
|
| 8374 |
+
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
|
| 8375 |
+
"mmlu_eval_accuracy_high_school_european_history": 0.5555555555555556,
|
| 8376 |
+
"mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
|
| 8377 |
+
"mmlu_eval_accuracy_high_school_government_and_politics": 0.6190476190476191,
|
| 8378 |
+
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5116279069767442,
|
| 8379 |
+
"mmlu_eval_accuracy_high_school_mathematics": 0.20689655172413793,
|
| 8380 |
+
"mmlu_eval_accuracy_high_school_microeconomics": 0.4230769230769231,
|
| 8381 |
+
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
|
| 8382 |
+
"mmlu_eval_accuracy_high_school_psychology": 0.8666666666666667,
|
| 8383 |
+
"mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
|
| 8384 |
+
"mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
|
| 8385 |
+
"mmlu_eval_accuracy_high_school_world_history": 0.6538461538461539,
|
| 8386 |
+
"mmlu_eval_accuracy_human_aging": 0.6956521739130435,
|
| 8387 |
+
"mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
|
| 8388 |
+
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
|
| 8389 |
+
"mmlu_eval_accuracy_jurisprudence": 0.36363636363636365,
|
| 8390 |
+
"mmlu_eval_accuracy_logical_fallacies": 0.6666666666666666,
|
| 8391 |
+
"mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
|
| 8392 |
+
"mmlu_eval_accuracy_management": 0.6363636363636364,
|
| 8393 |
+
"mmlu_eval_accuracy_marketing": 0.68,
|
| 8394 |
+
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
|
| 8395 |
+
"mmlu_eval_accuracy_miscellaneous": 0.6395348837209303,
|
| 8396 |
+
"mmlu_eval_accuracy_moral_disputes": 0.5,
|
| 8397 |
+
"mmlu_eval_accuracy_moral_scenarios": 0.27,
|
| 8398 |
+
"mmlu_eval_accuracy_nutrition": 0.6060606060606061,
|
| 8399 |
+
"mmlu_eval_accuracy_philosophy": 0.5,
|
| 8400 |
+
"mmlu_eval_accuracy_prehistory": 0.6285714285714286,
|
| 8401 |
+
"mmlu_eval_accuracy_professional_accounting": 0.3548387096774194,
|
| 8402 |
+
"mmlu_eval_accuracy_professional_law": 0.3588235294117647,
|
| 8403 |
+
"mmlu_eval_accuracy_professional_medicine": 0.5483870967741935,
|
| 8404 |
+
"mmlu_eval_accuracy_professional_psychology": 0.4927536231884058,
|
| 8405 |
+
"mmlu_eval_accuracy_public_relations": 0.5833333333333334,
|
| 8406 |
+
"mmlu_eval_accuracy_security_studies": 0.4444444444444444,
|
| 8407 |
+
"mmlu_eval_accuracy_sociology": 0.6363636363636364,
|
| 8408 |
+
"mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
|
| 8409 |
+
"mmlu_eval_accuracy_virology": 0.5,
|
| 8410 |
+
"mmlu_eval_accuracy_world_religions": 0.6842105263157895,
|
| 8411 |
+
"mmlu_loss": 1.3448165364505105,
|
| 8412 |
+
"step": 8800
|
| 8413 |
+
},
|
| 8414 |
+
{
|
| 8415 |
+
"epoch": 2.79,
|
| 8416 |
+
"learning_rate": 0.0002,
|
| 8417 |
+
"loss": 0.574,
|
| 8418 |
+
"step": 8810
|
| 8419 |
+
},
|
| 8420 |
+
{
|
| 8421 |
+
"epoch": 2.79,
|
| 8422 |
+
"learning_rate": 0.0002,
|
| 8423 |
+
"loss": 0.5965,
|
| 8424 |
+
"step": 8820
|
| 8425 |
+
},
|
| 8426 |
+
{
|
| 8427 |
+
"epoch": 2.8,
|
| 8428 |
+
"learning_rate": 0.0002,
|
| 8429 |
+
"loss": 0.5669,
|
| 8430 |
+
"step": 8830
|
| 8431 |
+
},
|
| 8432 |
+
{
|
| 8433 |
+
"epoch": 2.8,
|
| 8434 |
+
"learning_rate": 0.0002,
|
| 8435 |
+
"loss": 0.6088,
|
| 8436 |
+
"step": 8840
|
| 8437 |
+
},
|
| 8438 |
+
{
|
| 8439 |
+
"epoch": 2.8,
|
| 8440 |
+
"learning_rate": 0.0002,
|
| 8441 |
+
"loss": 0.542,
|
| 8442 |
+
"step": 8850
|
| 8443 |
+
},
|
| 8444 |
+
{
|
| 8445 |
+
"epoch": 2.81,
|
| 8446 |
+
"learning_rate": 0.0002,
|
| 8447 |
+
"loss": 0.5791,
|
| 8448 |
+
"step": 8860
|
| 8449 |
+
},
|
| 8450 |
+
{
|
| 8451 |
+
"epoch": 2.81,
|
| 8452 |
+
"learning_rate": 0.0002,
|
| 8453 |
+
"loss": 0.5663,
|
| 8454 |
+
"step": 8870
|
| 8455 |
+
},
|
| 8456 |
+
{
|
| 8457 |
+
"epoch": 2.81,
|
| 8458 |
+
"learning_rate": 0.0002,
|
| 8459 |
+
"loss": 0.6236,
|
| 8460 |
+
"step": 8880
|
| 8461 |
+
},
|
| 8462 |
+
{
|
| 8463 |
+
"epoch": 2.82,
|
| 8464 |
+
"learning_rate": 0.0002,
|
| 8465 |
+
"loss": 0.5634,
|
| 8466 |
+
"step": 8890
|
| 8467 |
+
},
|
| 8468 |
+
{
|
| 8469 |
+
"epoch": 2.82,
|
| 8470 |
+
"learning_rate": 0.0002,
|
| 8471 |
+
"loss": 0.5925,
|
| 8472 |
+
"step": 8900
|
| 8473 |
+
},
|
| 8474 |
+
{
|
| 8475 |
+
"epoch": 2.82,
|
| 8476 |
+
"learning_rate": 0.0002,
|
| 8477 |
+
"loss": 0.6357,
|
| 8478 |
+
"step": 8910
|
| 8479 |
+
},
|
| 8480 |
+
{
|
| 8481 |
+
"epoch": 2.83,
|
| 8482 |
+
"learning_rate": 0.0002,
|
| 8483 |
+
"loss": 0.5965,
|
| 8484 |
+
"step": 8920
|
| 8485 |
+
},
|
| 8486 |
+
{
|
| 8487 |
+
"epoch": 2.83,
|
| 8488 |
+
"learning_rate": 0.0002,
|
| 8489 |
+
"loss": 0.6071,
|
| 8490 |
+
"step": 8930
|
| 8491 |
+
},
|
| 8492 |
+
{
|
| 8493 |
+
"epoch": 2.83,
|
| 8494 |
+
"learning_rate": 0.0002,
|
| 8495 |
+
"loss": 0.6561,
|
| 8496 |
+
"step": 8940
|
| 8497 |
+
},
|
| 8498 |
+
{
|
| 8499 |
+
"epoch": 2.84,
|
| 8500 |
+
"learning_rate": 0.0002,
|
| 8501 |
+
"loss": 0.556,
|
| 8502 |
+
"step": 8950
|
| 8503 |
+
},
|
| 8504 |
+
{
|
| 8505 |
+
"epoch": 2.84,
|
| 8506 |
+
"learning_rate": 0.0002,
|
| 8507 |
+
"loss": 0.6129,
|
| 8508 |
+
"step": 8960
|
| 8509 |
+
},
|
| 8510 |
+
{
|
| 8511 |
+
"epoch": 2.84,
|
| 8512 |
+
"learning_rate": 0.0002,
|
| 8513 |
+
"loss": 0.535,
|
| 8514 |
+
"step": 8970
|
| 8515 |
+
},
|
| 8516 |
+
{
|
| 8517 |
+
"epoch": 2.85,
|
| 8518 |
+
"learning_rate": 0.0002,
|
| 8519 |
+
"loss": 0.6206,
|
| 8520 |
+
"step": 8980
|
| 8521 |
+
},
|
| 8522 |
+
{
|
| 8523 |
+
"epoch": 2.85,
|
| 8524 |
+
"learning_rate": 0.0002,
|
| 8525 |
+
"loss": 0.571,
|
| 8526 |
+
"step": 8990
|
| 8527 |
+
},
|
| 8528 |
+
{
|
| 8529 |
+
"epoch": 2.85,
|
| 8530 |
+
"learning_rate": 0.0002,
|
| 8531 |
+
"loss": 0.6166,
|
| 8532 |
+
"step": 9000
|
| 8533 |
+
},
|
| 8534 |
+
{
|
| 8535 |
+
"epoch": 2.85,
|
| 8536 |
+
"eval_loss": 0.7525665163993835,
|
| 8537 |
+
"eval_runtime": 111.1851,
|
| 8538 |
+
"eval_samples_per_second": 8.994,
|
| 8539 |
+
"eval_steps_per_second": 4.497,
|
| 8540 |
+
"step": 9000
|
| 8541 |
+
},
|
| 8542 |
+
{
|
| 8543 |
+
"epoch": 2.85,
|
| 8544 |
+
"mmlu_eval_accuracy": 0.485164659462464,
|
| 8545 |
+
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
|
| 8546 |
+
"mmlu_eval_accuracy_anatomy": 0.6428571428571429,
|
| 8547 |
+
"mmlu_eval_accuracy_astronomy": 0.375,
|
| 8548 |
+
"mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
|
| 8549 |
+
"mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
|
| 8550 |
+
"mmlu_eval_accuracy_college_biology": 0.375,
|
| 8551 |
+
"mmlu_eval_accuracy_college_chemistry": 0.25,
|
| 8552 |
+
"mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
|
| 8553 |
+
"mmlu_eval_accuracy_college_mathematics": 0.18181818181818182,
|
| 8554 |
+
"mmlu_eval_accuracy_college_medicine": 0.36363636363636365,
|
| 8555 |
+
"mmlu_eval_accuracy_college_physics": 0.18181818181818182,
|
| 8556 |
+
"mmlu_eval_accuracy_computer_security": 0.5454545454545454,
|
| 8557 |
+
"mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
|
| 8558 |
+
"mmlu_eval_accuracy_econometrics": 0.25,
|
| 8559 |
+
"mmlu_eval_accuracy_electrical_engineering": 0.3125,
|
| 8560 |
+
"mmlu_eval_accuracy_elementary_mathematics": 0.34146341463414637,
|
| 8561 |
+
"mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
|
| 8562 |
+
"mmlu_eval_accuracy_global_facts": 0.5,
|
| 8563 |
+
"mmlu_eval_accuracy_high_school_biology": 0.375,
|
| 8564 |
+
"mmlu_eval_accuracy_high_school_chemistry": 0.2727272727272727,
|
| 8565 |
+
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
|
| 8566 |
+
"mmlu_eval_accuracy_high_school_european_history": 0.6111111111111112,
|
| 8567 |
+
"mmlu_eval_accuracy_high_school_geography": 0.7272727272727273,
|
| 8568 |
+
"mmlu_eval_accuracy_high_school_government_and_politics": 0.5714285714285714,
|
| 8569 |
+
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5116279069767442,
|
| 8570 |
+
"mmlu_eval_accuracy_high_school_mathematics": 0.20689655172413793,
|
| 8571 |
+
"mmlu_eval_accuracy_high_school_microeconomics": 0.5384615384615384,
|
| 8572 |
+
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
|
| 8573 |
+
"mmlu_eval_accuracy_high_school_psychology": 0.8666666666666667,
|
| 8574 |
+
"mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
|
| 8575 |
+
"mmlu_eval_accuracy_high_school_us_history": 0.6363636363636364,
|
| 8576 |
+
"mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
|
| 8577 |
+
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
|
| 8578 |
+
"mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
|
| 8579 |
+
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
|
| 8580 |
+
"mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
|
| 8581 |
+
"mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
|
| 8582 |
+
"mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
|
| 8583 |
+
"mmlu_eval_accuracy_management": 0.6363636363636364,
|
| 8584 |
+
"mmlu_eval_accuracy_marketing": 0.64,
|
| 8585 |
+
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
|
| 8586 |
+
"mmlu_eval_accuracy_miscellaneous": 0.686046511627907,
|
| 8587 |
+
"mmlu_eval_accuracy_moral_disputes": 0.47368421052631576,
|
| 8588 |
+
"mmlu_eval_accuracy_moral_scenarios": 0.23,
|
| 8589 |
+
"mmlu_eval_accuracy_nutrition": 0.6060606060606061,
|
| 8590 |
+
"mmlu_eval_accuracy_philosophy": 0.5,
|
| 8591 |
+
"mmlu_eval_accuracy_prehistory": 0.5714285714285714,
|
| 8592 |
+
"mmlu_eval_accuracy_professional_accounting": 0.3548387096774194,
|
| 8593 |
+
"mmlu_eval_accuracy_professional_law": 0.32941176470588235,
|
| 8594 |
+
"mmlu_eval_accuracy_professional_medicine": 0.5483870967741935,
|
| 8595 |
+
"mmlu_eval_accuracy_professional_psychology": 0.4782608695652174,
|
| 8596 |
+
"mmlu_eval_accuracy_public_relations": 0.5,
|
| 8597 |
+
"mmlu_eval_accuracy_security_studies": 0.4444444444444444,
|
| 8598 |
+
"mmlu_eval_accuracy_sociology": 0.6818181818181818,
|
| 8599 |
+
"mmlu_eval_accuracy_us_foreign_policy": 0.8181818181818182,
|
| 8600 |
+
"mmlu_eval_accuracy_virology": 0.5,
|
| 8601 |
+
"mmlu_eval_accuracy_world_religions": 0.6842105263157895,
|
| 8602 |
+
"mmlu_loss": 1.3841694026952,
|
| 8603 |
+
"step": 9000
|
| 8604 |
+
},
|
| 8605 |
+
{
|
| 8606 |
+
"epoch": 2.85,
|
| 8607 |
+
"learning_rate": 0.0002,
|
| 8608 |
+
"loss": 0.553,
|
| 8609 |
+
"step": 9010
|
| 8610 |
+
},
|
| 8611 |
+
{
|
| 8612 |
+
"epoch": 2.86,
|
| 8613 |
+
"learning_rate": 0.0002,
|
| 8614 |
+
"loss": 0.572,
|
| 8615 |
+
"step": 9020
|
| 8616 |
+
},
|
| 8617 |
+
{
|
| 8618 |
+
"epoch": 2.86,
|
| 8619 |
+
"learning_rate": 0.0002,
|
| 8620 |
+
"loss": 0.6215,
|
| 8621 |
+
"step": 9030
|
| 8622 |
+
},
|
| 8623 |
+
{
|
| 8624 |
+
"epoch": 2.86,
|
| 8625 |
+
"learning_rate": 0.0002,
|
| 8626 |
+
"loss": 0.545,
|
| 8627 |
+
"step": 9040
|
| 8628 |
+
},
|
| 8629 |
+
{
|
| 8630 |
+
"epoch": 2.87,
|
| 8631 |
+
"learning_rate": 0.0002,
|
| 8632 |
+
"loss": 0.6241,
|
| 8633 |
+
"step": 9050
|
| 8634 |
+
},
|
| 8635 |
+
{
|
| 8636 |
+
"epoch": 2.87,
|
| 8637 |
+
"learning_rate": 0.0002,
|
| 8638 |
+
"loss": 0.5556,
|
| 8639 |
+
"step": 9060
|
| 8640 |
+
},
|
| 8641 |
+
{
|
| 8642 |
+
"epoch": 2.87,
|
| 8643 |
+
"learning_rate": 0.0002,
|
| 8644 |
+
"loss": 0.5995,
|
| 8645 |
+
"step": 9070
|
| 8646 |
+
},
|
| 8647 |
+
{
|
| 8648 |
+
"epoch": 2.88,
|
| 8649 |
+
"learning_rate": 0.0002,
|
| 8650 |
+
"loss": 0.6454,
|
| 8651 |
+
"step": 9080
|
| 8652 |
+
},
|
| 8653 |
+
{
|
| 8654 |
+
"epoch": 2.88,
|
| 8655 |
+
"learning_rate": 0.0002,
|
| 8656 |
+
"loss": 0.619,
|
| 8657 |
+
"step": 9090
|
| 8658 |
+
},
|
| 8659 |
+
{
|
| 8660 |
+
"epoch": 2.88,
|
| 8661 |
+
"learning_rate": 0.0002,
|
| 8662 |
+
"loss": 0.5746,
|
| 8663 |
+
"step": 9100
|
| 8664 |
+
},
|
| 8665 |
+
{
|
| 8666 |
+
"epoch": 2.89,
|
| 8667 |
+
"learning_rate": 0.0002,
|
| 8668 |
+
"loss": 0.6706,
|
| 8669 |
+
"step": 9110
|
| 8670 |
+
},
|
| 8671 |
+
{
|
| 8672 |
+
"epoch": 2.89,
|
| 8673 |
+
"learning_rate": 0.0002,
|
| 8674 |
+
"loss": 0.631,
|
| 8675 |
+
"step": 9120
|
| 8676 |
+
},
|
| 8677 |
+
{
|
| 8678 |
+
"epoch": 2.89,
|
| 8679 |
+
"learning_rate": 0.0002,
|
| 8680 |
+
"loss": 0.6514,
|
| 8681 |
+
"step": 9130
|
| 8682 |
+
},
|
| 8683 |
+
{
|
| 8684 |
+
"epoch": 2.9,
|
| 8685 |
+
"learning_rate": 0.0002,
|
| 8686 |
+
"loss": 0.6435,
|
| 8687 |
+
"step": 9140
|
| 8688 |
+
},
|
| 8689 |
+
{
|
| 8690 |
+
"epoch": 2.9,
|
| 8691 |
+
"learning_rate": 0.0002,
|
| 8692 |
+
"loss": 0.6059,
|
| 8693 |
+
"step": 9150
|
| 8694 |
+
},
|
| 8695 |
+
{
|
| 8696 |
+
"epoch": 2.9,
|
| 8697 |
+
"learning_rate": 0.0002,
|
| 8698 |
+
"loss": 0.5627,
|
| 8699 |
+
"step": 9160
|
| 8700 |
+
},
|
| 8701 |
+
{
|
| 8702 |
+
"epoch": 2.91,
|
| 8703 |
+
"learning_rate": 0.0002,
|
| 8704 |
+
"loss": 0.5996,
|
| 8705 |
+
"step": 9170
|
| 8706 |
+
},
|
| 8707 |
+
{
|
| 8708 |
+
"epoch": 2.91,
|
| 8709 |
+
"learning_rate": 0.0002,
|
| 8710 |
+
"loss": 0.73,
|
| 8711 |
+
"step": 9180
|
| 8712 |
+
},
|
| 8713 |
+
{
|
| 8714 |
+
"epoch": 2.91,
|
| 8715 |
+
"learning_rate": 0.0002,
|
| 8716 |
+
"loss": 0.6054,
|
| 8717 |
+
"step": 9190
|
| 8718 |
+
},
|
| 8719 |
+
{
|
| 8720 |
+
"epoch": 2.92,
|
| 8721 |
+
"learning_rate": 0.0002,
|
| 8722 |
+
"loss": 0.6257,
|
| 8723 |
+
"step": 9200
|
| 8724 |
+
},
|
| 8725 |
+
{
|
| 8726 |
+
"epoch": 2.92,
|
| 8727 |
+
"eval_loss": 0.7559297680854797,
|
| 8728 |
+
"eval_runtime": 111.3501,
|
| 8729 |
+
"eval_samples_per_second": 8.981,
|
| 8730 |
+
"eval_steps_per_second": 4.49,
|
| 8731 |
+
"step": 9200
|
| 8732 |
+
},
|
| 8733 |
+
{
|
| 8734 |
+
"epoch": 2.92,
|
| 8735 |
+
"mmlu_eval_accuracy": 0.46939903146057577,
|
| 8736 |
+
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
|
| 8737 |
+
"mmlu_eval_accuracy_anatomy": 0.6428571428571429,
|
| 8738 |
+
"mmlu_eval_accuracy_astronomy": 0.4375,
|
| 8739 |
+
"mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
|
| 8740 |
+
"mmlu_eval_accuracy_clinical_knowledge": 0.5172413793103449,
|
| 8741 |
+
"mmlu_eval_accuracy_college_biology": 0.375,
|
| 8742 |
+
"mmlu_eval_accuracy_college_chemistry": 0.25,
|
| 8743 |
+
"mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
|
| 8744 |
+
"mmlu_eval_accuracy_college_mathematics": 0.09090909090909091,
|
| 8745 |
+
"mmlu_eval_accuracy_college_medicine": 0.4090909090909091,
|
| 8746 |
+
"mmlu_eval_accuracy_college_physics": 0.18181818181818182,
|
| 8747 |
+
"mmlu_eval_accuracy_computer_security": 0.36363636363636365,
|
| 8748 |
+
"mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
|
| 8749 |
+
"mmlu_eval_accuracy_econometrics": 0.16666666666666666,
|
| 8750 |
+
"mmlu_eval_accuracy_electrical_engineering": 0.25,
|
| 8751 |
+
"mmlu_eval_accuracy_elementary_mathematics": 0.34146341463414637,
|
| 8752 |
+
"mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
|
| 8753 |
+
"mmlu_eval_accuracy_global_facts": 0.3,
|
| 8754 |
+
"mmlu_eval_accuracy_high_school_biology": 0.375,
|
| 8755 |
+
"mmlu_eval_accuracy_high_school_chemistry": 0.22727272727272727,
|
| 8756 |
+
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
|
| 8757 |
+
"mmlu_eval_accuracy_high_school_european_history": 0.5555555555555556,
|
| 8758 |
+
"mmlu_eval_accuracy_high_school_geography": 0.7272727272727273,
|
| 8759 |
+
"mmlu_eval_accuracy_high_school_government_and_politics": 0.6190476190476191,
|
| 8760 |
+
"mmlu_eval_accuracy_high_school_macroeconomics": 0.46511627906976744,
|
| 8761 |
+
"mmlu_eval_accuracy_high_school_mathematics": 0.20689655172413793,
|
| 8762 |
+
"mmlu_eval_accuracy_high_school_microeconomics": 0.4230769230769231,
|
| 8763 |
+
"mmlu_eval_accuracy_high_school_physics": 0.058823529411764705,
|
| 8764 |
+
"mmlu_eval_accuracy_high_school_psychology": 0.8333333333333334,
|
| 8765 |
+
"mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
|
| 8766 |
+
"mmlu_eval_accuracy_high_school_us_history": 0.6363636363636364,
|
| 8767 |
+
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
|
| 8768 |
+
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
|
| 8769 |
+
"mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
|
| 8770 |
+
"mmlu_eval_accuracy_international_law": 0.8461538461538461,
|
| 8771 |
+
"mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
|
| 8772 |
+
"mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
|
| 8773 |
+
"mmlu_eval_accuracy_machine_learning": 0.18181818181818182,
|
| 8774 |
+
"mmlu_eval_accuracy_management": 0.7272727272727273,
|
| 8775 |
+
"mmlu_eval_accuracy_marketing": 0.72,
|
| 8776 |
+
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
|
| 8777 |
+
"mmlu_eval_accuracy_miscellaneous": 0.6511627906976745,
|
| 8778 |
+
"mmlu_eval_accuracy_moral_disputes": 0.4473684210526316,
|
| 8779 |
+
"mmlu_eval_accuracy_moral_scenarios": 0.27,
|
| 8780 |
+
"mmlu_eval_accuracy_nutrition": 0.5757575757575758,
|
| 8781 |
+
"mmlu_eval_accuracy_philosophy": 0.5588235294117647,
|
| 8782 |
+
"mmlu_eval_accuracy_prehistory": 0.5714285714285714,
|
| 8783 |
+
"mmlu_eval_accuracy_professional_accounting": 0.3548387096774194,
|
| 8784 |
+
"mmlu_eval_accuracy_professional_law": 0.3176470588235294,
|
| 8785 |
+
"mmlu_eval_accuracy_professional_medicine": 0.5806451612903226,
|
| 8786 |
+
"mmlu_eval_accuracy_professional_psychology": 0.4927536231884058,
|
| 8787 |
+
"mmlu_eval_accuracy_public_relations": 0.5833333333333334,
|
| 8788 |
+
"mmlu_eval_accuracy_security_studies": 0.4444444444444444,
|
| 8789 |
+
"mmlu_eval_accuracy_sociology": 0.6363636363636364,
|
| 8790 |
+
"mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
|
| 8791 |
+
"mmlu_eval_accuracy_virology": 0.5,
|
| 8792 |
+
"mmlu_eval_accuracy_world_religions": 0.6842105263157895,
|
| 8793 |
+
"mmlu_loss": 1.4868696913628914,
|
| 8794 |
+
"step": 9200
|
| 8795 |
+
},
|
| 8796 |
+
{
|
| 8797 |
+
"epoch": 2.92,
|
| 8798 |
+
"learning_rate": 0.0002,
|
| 8799 |
+
"loss": 0.6179,
|
| 8800 |
+
"step": 9210
|
| 8801 |
+
},
|
| 8802 |
+
{
|
| 8803 |
+
"epoch": 2.92,
|
| 8804 |
+
"learning_rate": 0.0002,
|
| 8805 |
+
"loss": 0.5577,
|
| 8806 |
+
"step": 9220
|
| 8807 |
+
},
|
| 8808 |
+
{
|
| 8809 |
+
"epoch": 2.92,
|
| 8810 |
+
"learning_rate": 0.0002,
|
| 8811 |
+
"loss": 0.6043,
|
| 8812 |
+
"step": 9230
|
| 8813 |
+
},
|
| 8814 |
+
{
|
| 8815 |
+
"epoch": 2.93,
|
| 8816 |
+
"learning_rate": 0.0002,
|
| 8817 |
+
"loss": 0.6509,
|
| 8818 |
+
"step": 9240
|
| 8819 |
+
},
|
| 8820 |
+
{
|
| 8821 |
+
"epoch": 2.93,
|
| 8822 |
+
"learning_rate": 0.0002,
|
| 8823 |
+
"loss": 0.6184,
|
| 8824 |
+
"step": 9250
|
| 8825 |
+
},
|
| 8826 |
+
{
|
| 8827 |
+
"epoch": 2.93,
|
| 8828 |
+
"learning_rate": 0.0002,
|
| 8829 |
+
"loss": 0.6402,
|
| 8830 |
+
"step": 9260
|
| 8831 |
+
},
|
| 8832 |
+
{
|
| 8833 |
+
"epoch": 2.94,
|
| 8834 |
+
"learning_rate": 0.0002,
|
| 8835 |
+
"loss": 0.6614,
|
| 8836 |
+
"step": 9270
|
| 8837 |
+
},
|
| 8838 |
+
{
|
| 8839 |
+
"epoch": 2.94,
|
| 8840 |
+
"learning_rate": 0.0002,
|
| 8841 |
+
"loss": 0.6394,
|
| 8842 |
+
"step": 9280
|
| 8843 |
+
},
|
| 8844 |
+
{
|
| 8845 |
+
"epoch": 2.94,
|
| 8846 |
+
"learning_rate": 0.0002,
|
| 8847 |
+
"loss": 0.6085,
|
| 8848 |
+
"step": 9290
|
| 8849 |
+
},
|
| 8850 |
+
{
|
| 8851 |
+
"epoch": 2.95,
|
| 8852 |
+
"learning_rate": 0.0002,
|
| 8853 |
+
"loss": 0.6297,
|
| 8854 |
+
"step": 9300
|
| 8855 |
+
},
|
| 8856 |
+
{
|
| 8857 |
+
"epoch": 2.95,
|
| 8858 |
+
"learning_rate": 0.0002,
|
| 8859 |
+
"loss": 0.6099,
|
| 8860 |
+
"step": 9310
|
| 8861 |
+
},
|
| 8862 |
+
{
|
| 8863 |
+
"epoch": 2.95,
|
| 8864 |
+
"learning_rate": 0.0002,
|
| 8865 |
+
"loss": 0.6259,
|
| 8866 |
+
"step": 9320
|
| 8867 |
+
},
|
| 8868 |
+
{
|
| 8869 |
+
"epoch": 2.96,
|
| 8870 |
+
"learning_rate": 0.0002,
|
| 8871 |
+
"loss": 0.6226,
|
| 8872 |
+
"step": 9330
|
| 8873 |
+
},
|
| 8874 |
+
{
|
| 8875 |
+
"epoch": 2.96,
|
| 8876 |
+
"learning_rate": 0.0002,
|
| 8877 |
+
"loss": 0.6528,
|
| 8878 |
+
"step": 9340
|
| 8879 |
+
},
|
| 8880 |
+
{
|
| 8881 |
+
"epoch": 2.96,
|
| 8882 |
+
"learning_rate": 0.0002,
|
| 8883 |
+
"loss": 0.6388,
|
| 8884 |
+
"step": 9350
|
| 8885 |
+
},
|
| 8886 |
+
{
|
| 8887 |
+
"epoch": 2.97,
|
| 8888 |
+
"learning_rate": 0.0002,
|
| 8889 |
+
"loss": 0.6374,
|
| 8890 |
+
"step": 9360
|
| 8891 |
+
},
|
| 8892 |
+
{
|
| 8893 |
+
"epoch": 2.97,
|
| 8894 |
+
"learning_rate": 0.0002,
|
| 8895 |
+
"loss": 0.6096,
|
| 8896 |
+
"step": 9370
|
| 8897 |
+
},
|
| 8898 |
+
{
|
| 8899 |
+
"epoch": 2.97,
|
| 8900 |
+
"learning_rate": 0.0002,
|
| 8901 |
+
"loss": 0.598,
|
| 8902 |
+
"step": 9380
|
| 8903 |
+
},
|
| 8904 |
+
{
|
| 8905 |
+
"epoch": 2.98,
|
| 8906 |
+
"learning_rate": 0.0002,
|
| 8907 |
+
"loss": 0.5925,
|
| 8908 |
+
"step": 9390
|
| 8909 |
+
},
|
| 8910 |
+
{
|
| 8911 |
+
"epoch": 2.98,
|
| 8912 |
+
"learning_rate": 0.0002,
|
| 8913 |
+
"loss": 0.6067,
|
| 8914 |
+
"step": 9400
|
| 8915 |
+
},
|
| 8916 |
+
{
|
| 8917 |
+
"epoch": 2.98,
|
| 8918 |
+
"eval_loss": 0.754828929901123,
|
| 8919 |
+
"eval_runtime": 111.4146,
|
| 8920 |
+
"eval_samples_per_second": 8.975,
|
| 8921 |
+
"eval_steps_per_second": 4.488,
|
| 8922 |
+
"step": 9400
|
| 8923 |
+
},
|
| 8924 |
+
{
|
| 8925 |
+
"epoch": 2.98,
|
| 8926 |
+
"mmlu_eval_accuracy": 0.4664813810860653,
|
| 8927 |
+
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
|
| 8928 |
+
"mmlu_eval_accuracy_anatomy": 0.6428571428571429,
|
| 8929 |
+
"mmlu_eval_accuracy_astronomy": 0.4375,
|
| 8930 |
+
"mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
|
| 8931 |
+
"mmlu_eval_accuracy_clinical_knowledge": 0.5172413793103449,
|
| 8932 |
+
"mmlu_eval_accuracy_college_biology": 0.375,
|
| 8933 |
+
"mmlu_eval_accuracy_college_chemistry": 0.25,
|
| 8934 |
+
"mmlu_eval_accuracy_college_computer_science": 0.2727272727272727,
|
| 8935 |
+
"mmlu_eval_accuracy_college_mathematics": 0.09090909090909091,
|
| 8936 |
+
"mmlu_eval_accuracy_college_medicine": 0.36363636363636365,
|
| 8937 |
+
"mmlu_eval_accuracy_college_physics": 0.18181818181818182,
|
| 8938 |
+
"mmlu_eval_accuracy_computer_security": 0.45454545454545453,
|
| 8939 |
+
"mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
|
| 8940 |
+
"mmlu_eval_accuracy_econometrics": 0.16666666666666666,
|
| 8941 |
+
"mmlu_eval_accuracy_electrical_engineering": 0.3125,
|
| 8942 |
+
"mmlu_eval_accuracy_elementary_mathematics": 0.34146341463414637,
|
| 8943 |
+
"mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
|
| 8944 |
+
"mmlu_eval_accuracy_global_facts": 0.5,
|
| 8945 |
+
"mmlu_eval_accuracy_high_school_biology": 0.375,
|
| 8946 |
+
"mmlu_eval_accuracy_high_school_chemistry": 0.22727272727272727,
|
| 8947 |
+
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
|
| 8948 |
+
"mmlu_eval_accuracy_high_school_european_history": 0.6111111111111112,
|
| 8949 |
+
"mmlu_eval_accuracy_high_school_geography": 0.8636363636363636,
|
| 8950 |
+
"mmlu_eval_accuracy_high_school_government_and_politics": 0.5714285714285714,
|
| 8951 |
+
"mmlu_eval_accuracy_high_school_macroeconomics": 0.4883720930232558,
|
| 8952 |
+
"mmlu_eval_accuracy_high_school_mathematics": 0.20689655172413793,
|
| 8953 |
+
"mmlu_eval_accuracy_high_school_microeconomics": 0.5,
|
| 8954 |
+
"mmlu_eval_accuracy_high_school_physics": 0.058823529411764705,
|
| 8955 |
+
"mmlu_eval_accuracy_high_school_psychology": 0.85,
|
| 8956 |
+
"mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
|
| 8957 |
+
"mmlu_eval_accuracy_high_school_us_history": 0.6363636363636364,
|
| 8958 |
+
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
|
| 8959 |
+
"mmlu_eval_accuracy_human_aging": 0.6956521739130435,
|
| 8960 |
+
"mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
|
| 8961 |
+
"mmlu_eval_accuracy_international_law": 0.9230769230769231,
|
| 8962 |
+
"mmlu_eval_accuracy_jurisprudence": 0.36363636363636365,
|
| 8963 |
+
"mmlu_eval_accuracy_logical_fallacies": 0.5555555555555556,
|
| 8964 |
+
"mmlu_eval_accuracy_machine_learning": 0.18181818181818182,
|
| 8965 |
+
"mmlu_eval_accuracy_management": 0.6363636363636364,
|
| 8966 |
+
"mmlu_eval_accuracy_marketing": 0.68,
|
| 8967 |
+
"mmlu_eval_accuracy_medical_genetics": 0.8181818181818182,
|
| 8968 |
+
"mmlu_eval_accuracy_miscellaneous": 0.6511627906976745,
|
| 8969 |
+
"mmlu_eval_accuracy_moral_disputes": 0.5526315789473685,
|
| 8970 |
+
"mmlu_eval_accuracy_moral_scenarios": 0.27,
|
| 8971 |
+
"mmlu_eval_accuracy_nutrition": 0.5757575757575758,
|
| 8972 |
+
"mmlu_eval_accuracy_philosophy": 0.5,
|
| 8973 |
+
"mmlu_eval_accuracy_prehistory": 0.5142857142857142,
|
| 8974 |
+
"mmlu_eval_accuracy_professional_accounting": 0.3548387096774194,
|
| 8975 |
+
"mmlu_eval_accuracy_professional_law": 0.31176470588235294,
|
| 8976 |
+
"mmlu_eval_accuracy_professional_medicine": 0.5161290322580645,
|
| 8977 |
+
"mmlu_eval_accuracy_professional_psychology": 0.4927536231884058,
|
| 8978 |
+
"mmlu_eval_accuracy_public_relations": 0.5,
|
| 8979 |
+
"mmlu_eval_accuracy_security_studies": 0.4074074074074074,
|
| 8980 |
+
"mmlu_eval_accuracy_sociology": 0.6363636363636364,
|
| 8981 |
+
"mmlu_eval_accuracy_us_foreign_policy": 0.6363636363636364,
|
| 8982 |
+
"mmlu_eval_accuracy_virology": 0.4444444444444444,
|
| 8983 |
+
"mmlu_eval_accuracy_world_religions": 0.631578947368421,
|
| 8984 |
+
"mmlu_loss": 1.263844348665317,
|
| 8985 |
+
"step": 9400
|
| 8986 |
+
},
|
| 8987 |
+
{
|
| 8988 |
+
"epoch": 2.98,
|
| 8989 |
+
"learning_rate": 0.0002,
|
| 8990 |
+
"loss": 0.6571,
|
| 8991 |
+
"step": 9410
|
| 8992 |
+
},
|
| 8993 |
+
{
|
| 8994 |
+
"epoch": 2.98,
|
| 8995 |
+
"learning_rate": 0.0002,
|
| 8996 |
+
"loss": 0.6461,
|
| 8997 |
+
"step": 9420
|
| 8998 |
+
},
|
| 8999 |
+
{
|
| 9000 |
+
"epoch": 2.99,
|
| 9001 |
+
"learning_rate": 0.0002,
|
| 9002 |
+
"loss": 0.5997,
|
| 9003 |
+
"step": 9430
|
| 9004 |
+
},
|
| 9005 |
+
{
|
| 9006 |
+
"epoch": 2.99,
|
| 9007 |
+
"learning_rate": 0.0002,
|
| 9008 |
+
"loss": 0.5978,
|
| 9009 |
+
"step": 9440
|
| 9010 |
+
},
|
| 9011 |
+
{
|
| 9012 |
+
"epoch": 2.99,
|
| 9013 |
+
"learning_rate": 0.0002,
|
| 9014 |
+
"loss": 0.6302,
|
| 9015 |
+
"step": 9450
|
| 9016 |
+
},
|
| 9017 |
+
{
|
| 9018 |
+
"epoch": 3.0,
|
| 9019 |
+
"learning_rate": 0.0002,
|
| 9020 |
+
"loss": 0.5653,
|
| 9021 |
+
"step": 9460
|
| 9022 |
+
},
|
| 9023 |
+
{
|
| 9024 |
+
"epoch": 3.0,
|
| 9025 |
+
"learning_rate": 0.0002,
|
| 9026 |
+
"loss": 0.6311,
|
| 9027 |
+
"step": 9470
|
| 9028 |
+
},
|
| 9029 |
+
{
|
| 9030 |
+
"epoch": 3.0,
|
| 9031 |
+
"learning_rate": 0.0002,
|
| 9032 |
+
"loss": 0.5036,
|
| 9033 |
+
"step": 9480
|
| 9034 |
+
},
|
| 9035 |
+
{
|
| 9036 |
+
"epoch": 3.01,
|
| 9037 |
+
"learning_rate": 0.0002,
|
| 9038 |
+
"loss": 0.4517,
|
| 9039 |
+
"step": 9490
|
| 9040 |
+
},
|
| 9041 |
+
{
|
| 9042 |
+
"epoch": 3.01,
|
| 9043 |
+
"learning_rate": 0.0002,
|
| 9044 |
+
"loss": 0.4403,
|
| 9045 |
+
"step": 9500
|
| 9046 |
+
},
|
| 9047 |
+
{
|
| 9048 |
+
"epoch": 3.01,
|
| 9049 |
+
"learning_rate": 0.0002,
|
| 9050 |
+
"loss": 0.4921,
|
| 9051 |
+
"step": 9510
|
| 9052 |
+
},
|
| 9053 |
+
{
|
| 9054 |
+
"epoch": 3.02,
|
| 9055 |
+
"learning_rate": 0.0002,
|
| 9056 |
+
"loss": 0.464,
|
| 9057 |
+
"step": 9520
|
| 9058 |
+
},
|
| 9059 |
+
{
|
| 9060 |
+
"epoch": 3.02,
|
| 9061 |
+
"learning_rate": 0.0002,
|
| 9062 |
+
"loss": 0.4751,
|
| 9063 |
+
"step": 9530
|
| 9064 |
+
},
|
| 9065 |
+
{
|
| 9066 |
+
"epoch": 3.02,
|
| 9067 |
+
"learning_rate": 0.0002,
|
| 9068 |
+
"loss": 0.521,
|
| 9069 |
+
"step": 9540
|
| 9070 |
+
},
|
| 9071 |
+
{
|
| 9072 |
+
"epoch": 3.03,
|
| 9073 |
+
"learning_rate": 0.0002,
|
| 9074 |
+
"loss": 0.4261,
|
| 9075 |
+
"step": 9550
|
| 9076 |
+
},
|
| 9077 |
+
{
|
| 9078 |
+
"epoch": 3.03,
|
| 9079 |
+
"learning_rate": 0.0002,
|
| 9080 |
+
"loss": 0.4827,
|
| 9081 |
+
"step": 9560
|
| 9082 |
+
},
|
| 9083 |
+
{
|
| 9084 |
+
"epoch": 3.03,
|
| 9085 |
+
"learning_rate": 0.0002,
|
| 9086 |
+
"loss": 0.521,
|
| 9087 |
+
"step": 9570
|
| 9088 |
+
},
|
| 9089 |
+
{
|
| 9090 |
+
"epoch": 3.04,
|
| 9091 |
+
"learning_rate": 0.0002,
|
| 9092 |
+
"loss": 0.5273,
|
| 9093 |
+
"step": 9580
|
| 9094 |
+
},
|
| 9095 |
+
{
|
| 9096 |
+
"epoch": 3.04,
|
| 9097 |
+
"learning_rate": 0.0002,
|
| 9098 |
+
"loss": 0.5053,
|
| 9099 |
+
"step": 9590
|
| 9100 |
+
},
|
| 9101 |
+
{
|
| 9102 |
+
"epoch": 3.04,
|
| 9103 |
+
"learning_rate": 0.0002,
|
| 9104 |
+
"loss": 0.4598,
|
| 9105 |
+
"step": 9600
|
| 9106 |
+
},
|
| 9107 |
+
{
|
| 9108 |
+
"epoch": 3.04,
|
| 9109 |
+
"eval_loss": 0.8103159070014954,
|
| 9110 |
+
"eval_runtime": 111.1191,
|
| 9111 |
+
"eval_samples_per_second": 8.999,
|
| 9112 |
+
"eval_steps_per_second": 4.5,
|
| 9113 |
+
"step": 9600
|
| 9114 |
+
},
|
| 9115 |
+
{
|
| 9116 |
+
"epoch": 3.04,
|
| 9117 |
+
"mmlu_eval_accuracy": 0.4738733345661677,
|
| 9118 |
+
"mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
|
| 9119 |
+
"mmlu_eval_accuracy_anatomy": 0.5,
|
| 9120 |
+
"mmlu_eval_accuracy_astronomy": 0.375,
|
| 9121 |
+
"mmlu_eval_accuracy_business_ethics": 0.45454545454545453,
|
| 9122 |
+
"mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
|
| 9123 |
+
"mmlu_eval_accuracy_college_biology": 0.375,
|
| 9124 |
+
"mmlu_eval_accuracy_college_chemistry": 0.375,
|
| 9125 |
+
"mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
|
| 9126 |
+
"mmlu_eval_accuracy_college_mathematics": 0.09090909090909091,
|
| 9127 |
+
"mmlu_eval_accuracy_college_medicine": 0.36363636363636365,
|
| 9128 |
+
"mmlu_eval_accuracy_college_physics": 0.18181818181818182,
|
| 9129 |
+
"mmlu_eval_accuracy_computer_security": 0.2727272727272727,
|
| 9130 |
+
"mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
|
| 9131 |
+
"mmlu_eval_accuracy_econometrics": 0.25,
|
| 9132 |
+
"mmlu_eval_accuracy_electrical_engineering": 0.3125,
|
| 9133 |
+
"mmlu_eval_accuracy_elementary_mathematics": 0.3170731707317073,
|
| 9134 |
+
"mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
|
| 9135 |
+
"mmlu_eval_accuracy_global_facts": 0.5,
|
| 9136 |
+
"mmlu_eval_accuracy_high_school_biology": 0.375,
|
| 9137 |
+
"mmlu_eval_accuracy_high_school_chemistry": 0.2727272727272727,
|
| 9138 |
+
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
|
| 9139 |
+
"mmlu_eval_accuracy_high_school_european_history": 0.6111111111111112,
|
| 9140 |
+
"mmlu_eval_accuracy_high_school_geography": 0.7272727272727273,
|
| 9141 |
+
"mmlu_eval_accuracy_high_school_government_and_politics": 0.6190476190476191,
|
| 9142 |
+
"mmlu_eval_accuracy_high_school_macroeconomics": 0.4418604651162791,
|
| 9143 |
+
"mmlu_eval_accuracy_high_school_mathematics": 0.1724137931034483,
|
| 9144 |
+
"mmlu_eval_accuracy_high_school_microeconomics": 0.6153846153846154,
|
| 9145 |
+
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
|
| 9146 |
+
"mmlu_eval_accuracy_high_school_psychology": 0.85,
|
| 9147 |
+
"mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
|
| 9148 |
+
"mmlu_eval_accuracy_high_school_us_history": 0.6363636363636364,
|
| 9149 |
+
"mmlu_eval_accuracy_high_school_world_history": 0.6923076923076923,
|
| 9150 |
+
"mmlu_eval_accuracy_human_aging": 0.7391304347826086,
|
| 9151 |
+
"mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
|
| 9152 |
+
"mmlu_eval_accuracy_international_law": 1.0,
|
| 9153 |
+
"mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
|
| 9154 |
+
"mmlu_eval_accuracy_logical_fallacies": 0.5555555555555556,
|
| 9155 |
+
"mmlu_eval_accuracy_machine_learning": 0.0,
|
| 9156 |
+
"mmlu_eval_accuracy_management": 0.6363636363636364,
|
| 9157 |
+
"mmlu_eval_accuracy_marketing": 0.64,
|
| 9158 |
+
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
|
| 9159 |
+
"mmlu_eval_accuracy_miscellaneous": 0.686046511627907,
|
| 9160 |
+
"mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
|
| 9161 |
+
"mmlu_eval_accuracy_moral_scenarios": 0.24,
|
| 9162 |
+
"mmlu_eval_accuracy_nutrition": 0.5757575757575758,
|
| 9163 |
+
"mmlu_eval_accuracy_philosophy": 0.5,
|
| 9164 |
+
"mmlu_eval_accuracy_prehistory": 0.5428571428571428,
|
| 9165 |
+
"mmlu_eval_accuracy_professional_accounting": 0.3225806451612903,
|
| 9166 |
+
"mmlu_eval_accuracy_professional_law": 0.3235294117647059,
|
| 9167 |
+
"mmlu_eval_accuracy_professional_medicine": 0.5483870967741935,
|
| 9168 |
+
"mmlu_eval_accuracy_professional_psychology": 0.463768115942029,
|
| 9169 |
+
"mmlu_eval_accuracy_public_relations": 0.5,
|
| 9170 |
+
"mmlu_eval_accuracy_security_studies": 0.4074074074074074,
|
| 9171 |
+
"mmlu_eval_accuracy_sociology": 0.7727272727272727,
|
| 9172 |
+
"mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
|
| 9173 |
+
"mmlu_eval_accuracy_virology": 0.5,
|
| 9174 |
+
"mmlu_eval_accuracy_world_religions": 0.6842105263157895,
|
| 9175 |
+
"mmlu_loss": 1.2339814666551647,
|
| 9176 |
+
"step": 9600
|
| 9177 |
+
},
|
| 9178 |
+
{
|
| 9179 |
+
"epoch": 3.04,
|
| 9180 |
+
"learning_rate": 0.0002,
|
| 9181 |
+
"loss": 0.4979,
|
| 9182 |
+
"step": 9610
|
| 9183 |
+
},
|
| 9184 |
+
{
|
| 9185 |
+
"epoch": 3.05,
|
| 9186 |
+
"learning_rate": 0.0002,
|
| 9187 |
+
"loss": 0.4181,
|
| 9188 |
+
"step": 9620
|
| 9189 |
+
},
|
| 9190 |
+
{
|
| 9191 |
+
"epoch": 3.05,
|
| 9192 |
+
"learning_rate": 0.0002,
|
| 9193 |
+
"loss": 0.4513,
|
| 9194 |
+
"step": 9630
|
| 9195 |
+
},
|
| 9196 |
+
{
|
| 9197 |
+
"epoch": 3.05,
|
| 9198 |
+
"learning_rate": 0.0002,
|
| 9199 |
+
"loss": 0.5453,
|
| 9200 |
+
"step": 9640
|
| 9201 |
+
},
|
| 9202 |
+
{
|
| 9203 |
+
"epoch": 3.06,
|
| 9204 |
+
"learning_rate": 0.0002,
|
| 9205 |
+
"loss": 0.4922,
|
| 9206 |
+
"step": 9650
|
| 9207 |
+
},
|
| 9208 |
+
{
|
| 9209 |
+
"epoch": 3.06,
|
| 9210 |
+
"learning_rate": 0.0002,
|
| 9211 |
+
"loss": 0.5147,
|
| 9212 |
+
"step": 9660
|
| 9213 |
+
},
|
| 9214 |
+
{
|
| 9215 |
+
"epoch": 3.06,
|
| 9216 |
+
"learning_rate": 0.0002,
|
| 9217 |
+
"loss": 0.4696,
|
| 9218 |
+
"step": 9670
|
| 9219 |
+
},
|
| 9220 |
+
{
|
| 9221 |
+
"epoch": 3.07,
|
| 9222 |
+
"learning_rate": 0.0002,
|
| 9223 |
+
"loss": 0.4238,
|
| 9224 |
+
"step": 9680
|
| 9225 |
+
},
|
| 9226 |
+
{
|
| 9227 |
+
"epoch": 3.07,
|
| 9228 |
+
"learning_rate": 0.0002,
|
| 9229 |
+
"loss": 0.4443,
|
| 9230 |
+
"step": 9690
|
| 9231 |
+
},
|
| 9232 |
+
{
|
| 9233 |
+
"epoch": 3.07,
|
| 9234 |
+
"learning_rate": 0.0002,
|
| 9235 |
+
"loss": 0.4889,
|
| 9236 |
+
"step": 9700
|
| 9237 |
+
},
|
| 9238 |
+
{
|
| 9239 |
+
"epoch": 3.08,
|
| 9240 |
+
"learning_rate": 0.0002,
|
| 9241 |
+
"loss": 0.5475,
|
| 9242 |
+
"step": 9710
|
| 9243 |
+
},
|
| 9244 |
+
{
|
| 9245 |
+
"epoch": 3.08,
|
| 9246 |
+
"learning_rate": 0.0002,
|
| 9247 |
+
"loss": 0.4442,
|
| 9248 |
+
"step": 9720
|
| 9249 |
+
},
|
| 9250 |
+
{
|
| 9251 |
+
"epoch": 3.08,
|
| 9252 |
+
"learning_rate": 0.0002,
|
| 9253 |
+
"loss": 0.4865,
|
| 9254 |
+
"step": 9730
|
| 9255 |
+
},
|
| 9256 |
+
{
|
| 9257 |
+
"epoch": 3.09,
|
| 9258 |
+
"learning_rate": 0.0002,
|
| 9259 |
+
"loss": 0.5303,
|
| 9260 |
+
"step": 9740
|
| 9261 |
+
},
|
| 9262 |
+
{
|
| 9263 |
+
"epoch": 3.09,
|
| 9264 |
+
"learning_rate": 0.0002,
|
| 9265 |
+
"loss": 0.4883,
|
| 9266 |
+
"step": 9750
|
| 9267 |
+
},
|
| 9268 |
+
{
|
| 9269 |
+
"epoch": 3.09,
|
| 9270 |
+
"learning_rate": 0.0002,
|
| 9271 |
+
"loss": 0.4706,
|
| 9272 |
+
"step": 9760
|
| 9273 |
+
},
|
| 9274 |
+
{
|
| 9275 |
+
"epoch": 3.1,
|
| 9276 |
+
"learning_rate": 0.0002,
|
| 9277 |
+
"loss": 0.4861,
|
| 9278 |
+
"step": 9770
|
| 9279 |
+
},
|
| 9280 |
+
{
|
| 9281 |
+
"epoch": 3.1,
|
| 9282 |
+
"learning_rate": 0.0002,
|
| 9283 |
+
"loss": 0.4185,
|
| 9284 |
+
"step": 9780
|
| 9285 |
+
},
|
| 9286 |
+
{
|
| 9287 |
+
"epoch": 3.1,
|
| 9288 |
+
"learning_rate": 0.0002,
|
| 9289 |
+
"loss": 0.5567,
|
| 9290 |
+
"step": 9790
|
| 9291 |
+
},
|
| 9292 |
+
{
|
| 9293 |
+
"epoch": 3.11,
|
| 9294 |
+
"learning_rate": 0.0002,
|
| 9295 |
+
"loss": 0.5311,
|
| 9296 |
+
"step": 9800
|
| 9297 |
+
},
|
| 9298 |
+
{
|
| 9299 |
+
"epoch": 3.11,
|
| 9300 |
+
"eval_loss": 0.806504487991333,
|
| 9301 |
+
"eval_runtime": 111.202,
|
| 9302 |
+
"eval_samples_per_second": 8.993,
|
| 9303 |
+
"eval_steps_per_second": 4.496,
|
| 9304 |
+
"step": 9800
|
| 9305 |
+
},
|
| 9306 |
+
{
|
| 9307 |
+
"epoch": 3.11,
|
| 9308 |
+
"mmlu_eval_accuracy": 0.48250940137873616,
|
| 9309 |
+
"mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
|
| 9310 |
+
"mmlu_eval_accuracy_anatomy": 0.5714285714285714,
|
| 9311 |
+
"mmlu_eval_accuracy_astronomy": 0.4375,
|
| 9312 |
+
"mmlu_eval_accuracy_business_ethics": 0.45454545454545453,
|
| 9313 |
+
"mmlu_eval_accuracy_clinical_knowledge": 0.5517241379310345,
|
| 9314 |
+
"mmlu_eval_accuracy_college_biology": 0.375,
|
| 9315 |
+
"mmlu_eval_accuracy_college_chemistry": 0.375,
|
| 9316 |
+
"mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
|
| 9317 |
+
"mmlu_eval_accuracy_college_mathematics": 0.09090909090909091,
|
| 9318 |
+
"mmlu_eval_accuracy_college_medicine": 0.3181818181818182,
|
| 9319 |
+
"mmlu_eval_accuracy_college_physics": 0.2727272727272727,
|
| 9320 |
+
"mmlu_eval_accuracy_computer_security": 0.36363636363636365,
|
| 9321 |
+
"mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
|
| 9322 |
+
"mmlu_eval_accuracy_econometrics": 0.25,
|
| 9323 |
+
"mmlu_eval_accuracy_electrical_engineering": 0.3125,
|
| 9324 |
+
"mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
|
| 9325 |
+
"mmlu_eval_accuracy_formal_logic": 0.14285714285714285,
|
| 9326 |
+
"mmlu_eval_accuracy_global_facts": 0.5,
|
| 9327 |
+
"mmlu_eval_accuracy_high_school_biology": 0.375,
|
| 9328 |
+
"mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
|
| 9329 |
+
"mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
|
| 9330 |
+
"mmlu_eval_accuracy_high_school_european_history": 0.6111111111111112,
|
| 9331 |
+
"mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
|
| 9332 |
+
"mmlu_eval_accuracy_high_school_government_and_politics": 0.6190476190476191,
|
| 9333 |
+
"mmlu_eval_accuracy_high_school_macroeconomics": 0.5348837209302325,
|
| 9334 |
+
"mmlu_eval_accuracy_high_school_mathematics": 0.13793103448275862,
|
| 9335 |
+
"mmlu_eval_accuracy_high_school_microeconomics": 0.5384615384615384,
|
| 9336 |
+
"mmlu_eval_accuracy_high_school_physics": 0.11764705882352941,
|
| 9337 |
+
"mmlu_eval_accuracy_high_school_psychology": 0.8666666666666667,
|
| 9338 |
+
"mmlu_eval_accuracy_high_school_statistics": 0.391304347826087,
|
| 9339 |
+
"mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
|
| 9340 |
+
"mmlu_eval_accuracy_high_school_world_history": 0.7307692307692307,
|
| 9341 |
+
"mmlu_eval_accuracy_human_aging": 0.6956521739130435,
|
| 9342 |
+
"mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
|
| 9343 |
+
"mmlu_eval_accuracy_international_law": 0.8461538461538461,
|
| 9344 |
+
"mmlu_eval_accuracy_jurisprudence": 0.36363636363636365,
|
| 9345 |
+
"mmlu_eval_accuracy_logical_fallacies": 0.5555555555555556,
|
| 9346 |
+
"mmlu_eval_accuracy_machine_learning": 0.18181818181818182,
|
| 9347 |
+
"mmlu_eval_accuracy_management": 0.6363636363636364,
|
| 9348 |
+
"mmlu_eval_accuracy_marketing": 0.64,
|
| 9349 |
+
"mmlu_eval_accuracy_medical_genetics": 0.9090909090909091,
|
| 9350 |
+
"mmlu_eval_accuracy_miscellaneous": 0.6627906976744186,
|
| 9351 |
+
"mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
|
| 9352 |
+
"mmlu_eval_accuracy_moral_scenarios": 0.24,
|
| 9353 |
+
"mmlu_eval_accuracy_nutrition": 0.6060606060606061,
|
| 9354 |
+
"mmlu_eval_accuracy_philosophy": 0.5,
|
| 9355 |
+
"mmlu_eval_accuracy_prehistory": 0.5142857142857142,
|
| 9356 |
+
"mmlu_eval_accuracy_professional_accounting": 0.3225806451612903,
|
| 9357 |
+
"mmlu_eval_accuracy_professional_law": 0.3411764705882353,
|
| 9358 |
+
"mmlu_eval_accuracy_professional_medicine": 0.5483870967741935,
|
| 9359 |
+
"mmlu_eval_accuracy_professional_psychology": 0.4927536231884058,
|
| 9360 |
+
"mmlu_eval_accuracy_public_relations": 0.4166666666666667,
|
| 9361 |
+
"mmlu_eval_accuracy_security_studies": 0.4444444444444444,
|
| 9362 |
+
"mmlu_eval_accuracy_sociology": 0.7272727272727273,
|
| 9363 |
+
"mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
|
| 9364 |
+
"mmlu_eval_accuracy_virology": 0.5,
|
| 9365 |
+
"mmlu_eval_accuracy_world_religions": 0.7894736842105263,
|
| 9366 |
+
"mmlu_loss": 1.4220807970581728,
|
| 9367 |
+
"step": 9800
|
| 9368 |
}
|
| 9369 |
],
|
| 9370 |
"max_steps": 10000,
|
| 9371 |
"num_train_epochs": 4,
|
| 9372 |
+
"total_flos": 2.9756699225771704e+18,
|
| 9373 |
"trial_name": null,
|
| 9374 |
"trial_params": null
|
| 9375 |
}
|
{checkpoint-7800 → checkpoint-9800}/training_args.bin
RENAMED
|
File without changes
|