fkjsdahfjkasf
commited on
Commit
•
51fc8ca
1
Parent(s):
2089a85
fkjsdahfjkasf/Test-SFTLlama-3-8B
Browse files
README.md
CHANGED
@@ -40,7 +40,7 @@ More information needed
|
|
40 |
The following hyperparameters were used during training:
|
41 |
- learning_rate: 0.0002
|
42 |
- train_batch_size: 4
|
43 |
-
- eval_batch_size:
|
44 |
- seed: 1337
|
45 |
- gradient_accumulation_steps: 4
|
46 |
- total_train_batch_size: 16
|
|
|
40 |
The following hyperparameters were used during training:
|
41 |
- learning_rate: 0.0002
|
42 |
- train_batch_size: 4
|
43 |
+
- eval_batch_size: 4
|
44 |
- seed: 1337
|
45 |
- gradient_accumulation_steps: 4
|
46 |
- total_train_batch_size: 16
|
adapter_config.json
CHANGED
@@ -15,21 +15,21 @@
|
|
15 |
"megatron_config": null,
|
16 |
"megatron_core": "megatron.core",
|
17 |
"modules_to_save": [
|
18 |
-
"
|
19 |
-
"
|
20 |
],
|
21 |
"peft_type": "LORA",
|
22 |
"r": 16,
|
23 |
"rank_pattern": {},
|
24 |
"revision": "unsloth",
|
25 |
"target_modules": [
|
26 |
-
"o_proj",
|
27 |
-
"v_proj",
|
28 |
"q_proj",
|
|
|
|
|
|
|
29 |
"down_proj",
|
30 |
"up_proj",
|
31 |
-
"
|
32 |
-
"gate_proj"
|
33 |
],
|
34 |
"task_type": "CAUSAL_LM",
|
35 |
"use_dora": false,
|
|
|
15 |
"megatron_config": null,
|
16 |
"megatron_core": "megatron.core",
|
17 |
"modules_to_save": [
|
18 |
+
"embed_tokens",
|
19 |
+
"lm_head"
|
20 |
],
|
21 |
"peft_type": "LORA",
|
22 |
"r": 16,
|
23 |
"rank_pattern": {},
|
24 |
"revision": "unsloth",
|
25 |
"target_modules": [
|
|
|
|
|
26 |
"q_proj",
|
27 |
+
"k_proj",
|
28 |
+
"gate_proj",
|
29 |
+
"o_proj",
|
30 |
"down_proj",
|
31 |
"up_proj",
|
32 |
+
"v_proj"
|
|
|
33 |
],
|
34 |
"task_type": "CAUSAL_LM",
|
35 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4370526560
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4e9f5d5f3504664faf8ff0644f67875957495be6aa92402ba1fd27edaaf91938
|
3 |
size 4370526560
|
runs/Jun10_00-16-48_150-136-147-171/events.out.tfevents.1717978611.150-136-147-171.14731.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:155799e1ff24cbdd150030de973f84cb2518eff671058a9a57b8bcffb6e58b9a
|
3 |
+
size 5693
|
trainer_state.json
CHANGED
@@ -10,18 +10,18 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
-
"num_input_tokens_seen":
|
14 |
"step": 1,
|
15 |
-
"total_flos":
|
16 |
-
"train_loss":
|
17 |
-
"train_runtime":
|
18 |
-
"train_samples_per_second": 0.
|
19 |
-
"train_steps_per_second": 0.
|
20 |
}
|
21 |
],
|
22 |
"logging_steps": 5,
|
23 |
"max_steps": 1,
|
24 |
-
"num_input_tokens_seen":
|
25 |
"num_train_epochs": 1,
|
26 |
"save_steps": 50,
|
27 |
"stateful_callbacks": {
|
@@ -36,7 +36,7 @@
|
|
36 |
"attributes": {}
|
37 |
}
|
38 |
},
|
39 |
-
"total_flos":
|
40 |
"train_batch_size": 4,
|
41 |
"trial_name": null,
|
42 |
"trial_params": null
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
+
"num_input_tokens_seen": 131072,
|
14 |
"step": 1,
|
15 |
+
"total_flos": 6348239762620416.0,
|
16 |
+
"train_loss": 2.5150043964385986,
|
17 |
+
"train_runtime": 47.1012,
|
18 |
+
"train_samples_per_second": 0.34,
|
19 |
+
"train_steps_per_second": 0.021
|
20 |
}
|
21 |
],
|
22 |
"logging_steps": 5,
|
23 |
"max_steps": 1,
|
24 |
+
"num_input_tokens_seen": 131072,
|
25 |
"num_train_epochs": 1,
|
26 |
"save_steps": 50,
|
27 |
"stateful_callbacks": {
|
|
|
36 |
"attributes": {}
|
37 |
}
|
38 |
},
|
39 |
+
"total_flos": 6348239762620416.0,
|
40 |
"train_batch_size": 4,
|
41 |
"trial_name": null,
|
42 |
"trial_params": null
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5112
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a45267d968446bd97aaa0a3e2de91a0c88bffbaea2477221bd003801bd50f9d3
|
3 |
size 5112
|