Upload GemmaForCausalLM
Browse files- config.json +2 -1
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "/workspace/restore/finetuning-subnet-user/slouth_gemma/eval_dataset_s_cleaned_eacc_this_will_work/output_final_weights",
|
| 3 |
"architectures": [
|
| 4 |
"GemmaForCausalLM"
|
| 5 |
],
|
|
@@ -18,6 +18,7 @@
|
|
| 18 |
"num_hidden_layers": 18,
|
| 19 |
"num_key_value_heads": 1,
|
| 20 |
"pad_token_id": 0,
|
|
|
|
| 21 |
"rms_norm_eps": 1e-06,
|
| 22 |
"rope_scaling": null,
|
| 23 |
"rope_theta": 10000.0,
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "/workspace/restore/finetuning-subnet-user/eacc/eval_dataset_opportunities_cleaned_/workspace/restore/finetuning-subnet-user/slouth_gemma/eval_dataset_s_cleaned_eacc_this_will_work/output_final_weights_OwOpeepeepoopoo_gemmerica_eacc_this_will_work_2/output_final_weights",
|
| 3 |
"architectures": [
|
| 4 |
"GemmaForCausalLM"
|
| 5 |
],
|
|
|
|
| 18 |
"num_hidden_layers": 18,
|
| 19 |
"num_key_value_heads": 1,
|
| 20 |
"pad_token_id": 0,
|
| 21 |
+
"pretraining_tp": 1,
|
| 22 |
"rms_norm_eps": 1e-06,
|
| 23 |
"rope_scaling": null,
|
| 24 |
"rope_theta": 10000.0,
|