Hugging Face's logo

CMU-AIR2
/
math-deepseek_FULL_HardArith_Interm

Text Generation
Transformers
Safetensors
llama
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
jdannem6 commited on
Commit
6a449e1
•
1 Parent(s): bccc236

Uploaded checkpoint-22500

Browse files
Files changed (7) hide show
  1. config.json +2 -2
  2. model.safetensors +3 -0
  3. optimizer.pt +2 -2
  4. rng_state.pth +1 -1
  5. scheduler.pt +1 -1
  6. trainer_state.json +0 -0
  7. training_args.bin +2 -2
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "deepseek-ai/deepseek-coder-1.3b-instruct",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
@@ -24,7 +24,7 @@
24
  },
25
  "rope_theta": 100000,
26
  "tie_word_embeddings": false,
27
- "torch_dtype": "float32",
28
  "transformers_version": "4.38.2",
29
  "use_cache": true,
30
  "vocab_size": 32256
 
1
  {
2
+ "_name_or_path": "CMU-AIR2/math-deepseek_FULL_HardArith_Interm",
3
  "architectures": [
4
  "LlamaForCausalLM"
5
  ],
 
24
  },
25
  "rope_theta": 100000,
26
  "tie_word_embeddings": false,
27
+ "torch_dtype": "bfloat16",
28
  "transformers_version": "4.38.2",
29
  "use_cache": true,
30
  "vocab_size": 32256
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b2abe072ec787b02808a3e11292ada4972d5522179a8017c0bea930a924e665
3
+ size 2692969128
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f6c0438b7ed3e4c55f24fa3150fa982da8011360f12c7b707448e8db4ad3f902
3
- size 2699039674
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6446d399ce5a9716faf63c24be8e20ec8bb8bd984db93a4c6d4137f16a9dbe2e
3
+ size 5386075202
rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:97b0f27722247239511c0d33808326948f3078297be0bebdd0214846f3609f1e
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c92a95a97d689d636b085d406167a1d143dce26fb83ee64d21cf4b37a120302
3
  size 14244
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:29c7a79b53a589de48d3b7a21df9c0d024be4dea79f68869f72fdc01ae3b212a
3
  size 1064
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d71c00e9bfbac2252002b6eca4a38910300bb6c14e6c56273842dfbc024260d9
3
  size 1064
trainer_state.json CHANGED
The diff for this file is too large to render. See raw diff
 
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:58f96cee7b3928934c63b8523b5cd7bdf09666d503f89f715c81697ecd9b70dd
3
- size 5048
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63e1f8c1d21e32727e1cac707dced41ab69c27be3684ca827ad6ac3e456a60df
3
+ size 4984