69f75c7c1e29b2a0688ccc6210d7dd3d70e33a57f44a8c4b6ea62647cd95c9ef
Browse files- config.json +1 -1
- plots.png +0 -0
- smash_config.json +1 -1
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "/tmp/
|
| 3 |
"activation_function": "gelu_new",
|
| 4 |
"architectures": [
|
| 5 |
"GPTNeoForCausalLM"
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "/tmp/tmpz7jmx1ri",
|
| 3 |
"activation_function": "gelu_new",
|
| 4 |
"architectures": [
|
| 5 |
"GPTNeoForCausalLM"
|
plots.png
CHANGED
|
|
smash_config.json
CHANGED
|
@@ -8,7 +8,7 @@
|
|
| 8 |
"compilers": "None",
|
| 9 |
"task": "text_text_generation",
|
| 10 |
"device": "cuda",
|
| 11 |
-
"cache_dir": "/ceph/hdd/staff/charpent/.cache/
|
| 12 |
"batch_size": 1,
|
| 13 |
"model_name": "roneneldan/TinyStories-8M",
|
| 14 |
"pruning_ratio": 0.0,
|
|
|
|
| 8 |
"compilers": "None",
|
| 9 |
"task": "text_text_generation",
|
| 10 |
"device": "cuda",
|
| 11 |
+
"cache_dir": "/ceph/hdd/staff/charpent/.cache/modelsdfsbubcw",
|
| 12 |
"batch_size": 1,
|
| 13 |
"model_name": "roneneldan/TinyStories-8M",
|
| 14 |
"pruning_ratio": 0.0,
|