Hugging Face's logo

donaldyy
/
gemma-2-2B-it-thinking-function_calling-V0

Transformers
TensorBoard
Safetensors
Generated from Trainer
trl
sft
Model card Files
xet
Metrics Community
donaldyy commited on
Commit
a3be174
·
verified ·
1 Parent(s): a54cf80

donaldyy/gemma-2-2B-it-thinking-function_calling-V0

Browse files
Files changed (12) hide show
  1. README.md +5 -5
  2. adapter_config.json +6 -11
  3. adapter_model.safetensors +2 -2
  4. added_tokens.json +10 -0
  5. runs/Apr09_21-32-33_donald-MS-7D50/events.out.tfevents.1744230760.donald-MS-7D50.90139.0 +3 -0
  6. runs/Apr10_22-14-16_donald-MS-7D50/events.out.tfevents.1744319657.donald-MS-7D50.62645.0 +3 -0
  7. runs/Apr10_22-18-45_donald-MS-7D50/events.out.tfevents.1744319927.donald-MS-7D50.63329.0 +3 -0
  8. runs/Apr10_22-19-28_donald-MS-7D50/events.out.tfevents.1744319970.donald-MS-7D50.63496.0 +3 -0
  9. runs/Apr10_22-20-29_donald-MS-7D50/events.out.tfevents.1744320030.donald-MS-7D50.63749.0 +3 -0
  10. runs/Apr10_22-27-34_donald-MS-7D50/events.out.tfevents.1744320455.donald-MS-7D50.64735.0 +3 -0
  11. tokenizer.model +3 -0
  12. training_args.bin +2 -2
README.md CHANGED
@@ -34,11 +34,11 @@ This model was trained with SFT.
34
 
35
  ### Framework versions
36
 
37
- - TRL: 0.15.1
38
- - Transformers: 4.48.2
39
- - Pytorch: 2.5.1
40
- - Datasets: 3.3.2
41
- - Tokenizers: 0.21.0
42
 
43
  ## Citations
44
 
 
34
 
35
  ### Framework versions
36
 
37
+ - TRL: 0.16.0
38
+ - Transformers: 4.51.2
39
+ - Pytorch: 2.6.0
40
+ - Datasets: 3.5.0
41
+ - Tokenizers: 0.21.1
42
 
43
  ## Citations
44
 
adapter_config.json CHANGED
@@ -3,17 +3,13 @@
3
  "auto_mapping": null,
4
  "base_model_name_or_path": "google/gemma-2-2b-it",
5
  "bias": "none",
6
- "eva_config": null,
7
- "exclude_modules": null,
8
  "fan_in_fan_out": false,
9
  "inference_mode": true,
10
  "init_lora_weights": true,
11
- "layer_replication": null,
12
  "layers_pattern": null,
13
  "layers_to_transform": null,
14
  "loftq_config": {},
15
  "lora_alpha": 64,
16
- "lora_bias": false,
17
  "lora_dropout": 0.05,
18
  "megatron_config": null,
19
  "megatron_core": "megatron.core",
@@ -24,16 +20,15 @@
24
  "revision": null,
25
  "target_modules": [
26
  "o_proj",
27
- "up_proj",
28
- "down_proj",
29
- "lm_head",
30
  "k_proj",
31
- "q_proj",
32
  "embed_tokens",
33
- "v_proj",
34
- "gate_proj"
 
 
 
35
  ],
36
  "task_type": "CAUSAL_LM",
37
- "use_dora": false,
38
  "use_rslora": false
39
  }
 
3
  "auto_mapping": null,
4
  "base_model_name_or_path": "google/gemma-2-2b-it",
5
  "bias": "none",
 
 
6
  "fan_in_fan_out": false,
7
  "inference_mode": true,
8
  "init_lora_weights": true,
 
9
  "layers_pattern": null,
10
  "layers_to_transform": null,
11
  "loftq_config": {},
12
  "lora_alpha": 64,
 
13
  "lora_dropout": 0.05,
14
  "megatron_config": null,
15
  "megatron_core": "megatron.core",
 
20
  "revision": null,
21
  "target_modules": [
22
  "o_proj",
23
+ "v_proj",
 
 
24
  "k_proj",
 
25
  "embed_tokens",
26
+ "lm_head",
27
+ "q_proj",
28
+ "down_proj",
29
+ "gate_proj",
30
+ "up_proj"
31
  ],
32
  "task_type": "CAUSAL_LM",
 
33
  "use_rslora": false
34
  }
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:97377ec25605a7ed026cff5cac8e9be46b69a9d1fa261f1203a9ad8a7a72a594
3
- size 2475549872
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b332c77a572d7a7b9fe5d8ff4c52b148cad4c4974b087a194606e11b7a812b8c
3
+ size 2417486168
added_tokens.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</think>": 256003,
3
+ "</tool_call>": 256005,
4
+ "</tool_reponse>": 256007,
5
+ "</tools>": 256001,
6
+ "<think>": 256002,
7
+ "<tool_call>": 256004,
8
+ "<tool_reponse>": 256006,
9
+ "<tools>": 256000
10
+ }
runs/Apr09_21-32-33_donald-MS-7D50/events.out.tfevents.1744230760.donald-MS-7D50.90139.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dde6db337e8ff7229415ba0e848afcb0827144ba75565cbb3e7ba693df5e0e2e
3
+ size 40533
runs/Apr10_22-14-16_donald-MS-7D50/events.out.tfevents.1744319657.donald-MS-7D50.62645.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4eae00a052e8155b60d5c2356d3d27e9a03833b85cc76b07b0d7dd6b4893dd04
3
+ size 5703
runs/Apr10_22-18-45_donald-MS-7D50/events.out.tfevents.1744319927.donald-MS-7D50.63329.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d435666a53297ee0ef0599317a33fc17aadac88019ede7396502275ff75c583
3
+ size 6351
runs/Apr10_22-19-28_donald-MS-7D50/events.out.tfevents.1744319970.donald-MS-7D50.63496.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6f40fb670f7810e4980daaefd7d1f7a879ff4b85a21d2b74436f7b0460b23cb7
3
+ size 5703
runs/Apr10_22-20-29_donald-MS-7D50/events.out.tfevents.1744320030.donald-MS-7D50.63749.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d4eb95d1d2b7c9cc5cd96a1bffbe9f27f73945acb41c7c813ab578d7fd62574
3
+ size 8945
runs/Apr10_22-27-34_donald-MS-7D50/events.out.tfevents.1744320455.donald-MS-7D50.64735.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1bf3a642b7a68f161965f6942634c14339f0fa24fff031e227a5274dfbf746f
3
+ size 9707
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61a7b147390c64585d6c3543dd6fc636906c9af3865a5548f27f31aee1d4c8e2
3
+ size 4241003
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f4b68b9b67232ea60ae45f3f71eba553f72558706a16274b3d7fdad563263910
3
- size 5624
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3efa2287e58a3cd281c25af4433eaf75692a4a3919ae9b1f823f2d190481a87d
3
+ size 5688