Hugging Face's logo

BTGFM
/
phi3.5-mini-adapter_v0

PEFT
TensorBoard
Safetensors
Generated from Trainer
Model card Files Metrics Community
BTGFM commited on
Commit
663a8e4
•
1 Parent(s): 8fa3744

Training in progress, step 20

Browse files
Files changed (7) hide show
  1. adapter_config.json +1 -1
  2. adapter_model.safetensors +1 -1
  3. runs/Aug21_13-52-07_B650E/events.out.tfevents.1724262738.B650E.11078.0 +3 -0
  4. runs/Aug21_13-57-05_B650E/events.out.tfevents.1724263031.B650E.11724.0 +3 -0
  5. runs/Aug21_14-12-58_B650E/events.out.tfevents.1724264208.B650E.14325.0 +3 -0
  6. train_args.json +30 -0
  7. training_args.bin +1 -1
adapter_config.json CHANGED
@@ -20,9 +20,9 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
- "qkv_proj",
24
  "gate_up_proj",
25
  "down_proj",
 
26
  "o_proj"
27
  ],
28
  "task_type": "CAUSAL_LM",
 
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
 
23
  "gate_up_proj",
24
  "down_proj",
25
+ "qkv_proj",
26
  "o_proj"
27
  ],
28
  "task_type": "CAUSAL_LM",
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6ceab4d46a606e222e8aad66faf1cec9196b2fa49f5e99fe94f5c1dde13cdbeb
3
  size 100697728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b41dbfc268f977616f68bd00d472a1f8c2b902b8c4574dfa66d0de3e9157063
3
  size 100697728
runs/Aug21_13-52-07_B650E/events.out.tfevents.1724262738.B650E.11078.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cddef292310720c37fe2f6a7b61b80d7423ae62487dd1294a266e794608d8384
3
+ size 8958
runs/Aug21_13-57-05_B650E/events.out.tfevents.1724263031.B650E.11724.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d82b9b85d78987139db7506c035525b4237d764313f04ef2b2cea6145cbb6936
3
+ size 9993
runs/Aug21_14-12-58_B650E/events.out.tfevents.1724264208.B650E.14325.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:335d3039a558cb69db12c23c610a86870670cc49c2260e6dbb1857cf92b5ccc2
3
+ size 13009
train_args.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "BASE_MODEL": "microsoft/Phi-3.5-mini-instruct",
3
+ "SEQ_LENGTH": 512,
4
+ "MAX_STEPS": 250,
5
+ "BATCH_SIZE": 4,
6
+ "GR_ACC_STEPS": 4,
7
+ "LR": 0.0002,
8
+ "LR_SCHEDULER_TYPE": "cosine",
9
+ "OPTIMIZER": "adamw_torch",
10
+ "WEIGHT_DECAY": 0.001,
11
+ "WARMUP_RATIO": 0.05,
12
+ "EVAL_FREQ": 10,
13
+ "SAVE_FREQ": 20,
14
+ "SAVE_LIMIT": 2,
15
+ "LOG_FREQ": 1,
16
+ "BF16": true,
17
+ "FP16": false,
18
+ "FIM_RATE": 0.5,
19
+ "FIM_SPM_RAT": 0.5,
20
+ "LORA_R": 16,
21
+ "LORA_ALPHA": 48,
22
+ "LORA_DROPOUT": 0.0,
23
+ "LORA_TARGET_MODULES": "all-linear",
24
+ "USE_NESTED_QUANT": true,
25
+ "BNB_4BIT_COMPUTE_DTYPE": "bfloat16",
26
+ "load_in_8bit": true,
27
+ "SEED": 0,
28
+ "EARLY_STOP_PATIENCE": 3,
29
+ "EARLY_STOP_THRESHOLD": 0.01
30
+ }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ec44a5f4553a62c475aa10ccd5d67ebc7d6a90b4ebf0d53848e1b2a467c6b898
3
  size 5240
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1024b1fced171dbdcc548744e15886402d1da3c41af9576db19aeedd489e34c7
3
  size 5240