scottsuk0306
commited on
Commit
•
5fb6fcd
1
Parent(s):
646ecdd
Model save
Browse files- README.md +7 -12
- all_results.json +5 -10
- config.json +1 -1
- generation_config.json +1 -1
- model-00001-of-00002.safetensors +1 -1
- model-00002-of-00002.safetensors +1 -1
- train_results.json +5 -5
- trainer_state.json +20 -20
- training_args.bin +2 -2
README.md
CHANGED
@@ -2,29 +2,24 @@
|
|
2 |
license: gemma
|
3 |
base_model: google/gemma-2-2b
|
4 |
tags:
|
5 |
-
- easylm
|
6 |
-
- trl
|
7 |
-
- sft
|
8 |
-
- generated_from_trainer
|
9 |
- trl
|
10 |
- sft
|
11 |
- generated_from_trainer
|
12 |
datasets:
|
13 |
- alpaca_farm
|
14 |
model-index:
|
15 |
-
- name: easylm-sft-gemma-2-2b
|
16 |
results: []
|
17 |
---
|
18 |
|
19 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
20 |
should probably proofread and complete it, then remove this comment. -->
|
21 |
|
22 |
-
|
23 |
-
# easylm-sft-gemma-2-2b-v2.7
|
24 |
|
25 |
This model is a fine-tuned version of [google/gemma-2-2b](https://huggingface.co/google/gemma-2-2b) on the alpaca_farm dataset.
|
26 |
It achieves the following results on the evaluation set:
|
27 |
-
- Loss: 0.
|
28 |
|
29 |
## Model description
|
30 |
|
@@ -59,14 +54,14 @@ The following hyperparameters were used during training:
|
|
59 |
|
60 |
| Training Loss | Epoch | Step | Validation Loss |
|
61 |
|:-------------:|:-----:|:----:|:---------------:|
|
62 |
-
| No log | 1.0 | 13 | 0.
|
63 |
-
| No log | 2.0 | 26 | 0.
|
64 |
-
| No log | 3.0 | 39 | 0.
|
65 |
|
66 |
|
67 |
### Framework versions
|
68 |
|
69 |
-
- Transformers 4.
|
70 |
- Pytorch 2.3.0+cu121
|
71 |
- Datasets 2.20.0
|
72 |
- Tokenizers 0.19.1
|
|
|
2 |
license: gemma
|
3 |
base_model: google/gemma-2-2b
|
4 |
tags:
|
|
|
|
|
|
|
|
|
5 |
- trl
|
6 |
- sft
|
7 |
- generated_from_trainer
|
8 |
datasets:
|
9 |
- alpaca_farm
|
10 |
model-index:
|
11 |
+
- name: easylm-sft-gemma-2-2b
|
12 |
results: []
|
13 |
---
|
14 |
|
15 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
16 |
should probably proofread and complete it, then remove this comment. -->
|
17 |
|
18 |
+
# easylm-sft-gemma-2-2b
|
|
|
19 |
|
20 |
This model is a fine-tuned version of [google/gemma-2-2b](https://huggingface.co/google/gemma-2-2b) on the alpaca_farm dataset.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.9477
|
23 |
|
24 |
## Model description
|
25 |
|
|
|
54 |
|
55 |
| Training Loss | Epoch | Step | Validation Loss |
|
56 |
|:-------------:|:-----:|:----:|:---------------:|
|
57 |
+
| No log | 1.0 | 13 | 0.7365 |
|
58 |
+
| No log | 2.0 | 26 | 0.9138 |
|
59 |
+
| No log | 3.0 | 39 | 0.9477 |
|
60 |
|
61 |
|
62 |
### Framework versions
|
63 |
|
64 |
+
- Transformers 4.43.3
|
65 |
- Pytorch 2.3.0+cu121
|
66 |
- Datasets 2.20.0
|
67 |
- Tokenizers 0.19.1
|
all_results.json
CHANGED
@@ -1,14 +1,9 @@
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
-
"
|
4 |
-
"
|
5 |
-
"
|
6 |
-
"eval_samples_per_second": 140.76,
|
7 |
-
"eval_steps_per_second": 18.299,
|
8 |
-
"total_flos": 648728419500032.0,
|
9 |
-
"train_loss": 0.4328929705497546,
|
10 |
-
"train_runtime": 169.1685,
|
11 |
"train_samples": 100,
|
12 |
-
"train_samples_per_second": 1.
|
13 |
-
"train_steps_per_second": 0.
|
14 |
}
|
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
+
"total_flos": 648728369168384.0,
|
4 |
+
"train_loss": 0.4277030749198718,
|
5 |
+
"train_runtime": 188.9615,
|
|
|
|
|
|
|
|
|
|
|
6 |
"train_samples": 100,
|
7 |
+
"train_samples_per_second": 1.588,
|
8 |
+
"train_steps_per_second": 0.206
|
9 |
}
|
config.json
CHANGED
@@ -27,7 +27,7 @@
|
|
27 |
"rope_theta": 10000.0,
|
28 |
"sliding_window": 4096,
|
29 |
"torch_dtype": "bfloat16",
|
30 |
-
"transformers_version": "4.
|
31 |
"use_cache": true,
|
32 |
"vocab_size": 256000
|
33 |
}
|
|
|
27 |
"rope_theta": 10000.0,
|
28 |
"sliding_window": 4096,
|
29 |
"torch_dtype": "bfloat16",
|
30 |
+
"transformers_version": "4.43.3",
|
31 |
"use_cache": true,
|
32 |
"vocab_size": 256000
|
33 |
}
|
generation_config.json
CHANGED
@@ -4,5 +4,5 @@
|
|
4 |
"cache_implementation": "hybrid",
|
5 |
"eos_token_id": 1,
|
6 |
"pad_token_id": 0,
|
7 |
-
"transformers_version": "4.
|
8 |
}
|
|
|
4 |
"cache_implementation": "hybrid",
|
5 |
"eos_token_id": 1,
|
6 |
"pad_token_id": 0,
|
7 |
+
"transformers_version": "4.43.3"
|
8 |
}
|
model-00001-of-00002.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4988025760
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5f939cd2fc3072d9610787f59518cfa616d9a4e4574d91879c09278b25c235f4
|
3 |
size 4988025760
|
model-00002-of-00002.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1420339880
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:39deb869e669d42dcaaa5dd503f2130a86ee9f294a1f1c6fcac4152fe9ee7ead
|
3 |
size 1420339880
|
train_results.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
-
"total_flos":
|
4 |
-
"train_loss": 0.
|
5 |
-
"train_runtime":
|
6 |
"train_samples": 100,
|
7 |
-
"train_samples_per_second": 1.
|
8 |
-
"train_steps_per_second": 0.
|
9 |
}
|
|
|
1 |
{
|
2 |
"epoch": 3.0,
|
3 |
+
"total_flos": 648728369168384.0,
|
4 |
+
"train_loss": 0.4277030749198718,
|
5 |
+
"train_runtime": 188.9615,
|
6 |
"train_samples": 100,
|
7 |
+
"train_samples_per_second": 1.588,
|
8 |
+
"train_steps_per_second": 0.206
|
9 |
}
|
trainer_state.json
CHANGED
@@ -10,36 +10,36 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
-
"eval_loss": 0.
|
14 |
-
"eval_runtime": 0.
|
15 |
-
"eval_samples_per_second":
|
16 |
-
"eval_steps_per_second": 18.
|
17 |
"step": 13
|
18 |
},
|
19 |
{
|
20 |
"epoch": 2.0,
|
21 |
-
"eval_loss": 0.
|
22 |
-
"eval_runtime": 0.
|
23 |
-
"eval_samples_per_second": 143.
|
24 |
-
"eval_steps_per_second": 18.
|
25 |
"step": 26
|
26 |
},
|
27 |
{
|
28 |
"epoch": 3.0,
|
29 |
-
"eval_loss": 0.
|
30 |
-
"eval_runtime": 0.
|
31 |
-
"eval_samples_per_second":
|
32 |
-
"eval_steps_per_second": 18.
|
33 |
"step": 39
|
34 |
},
|
35 |
{
|
36 |
"epoch": 3.0,
|
37 |
"step": 39,
|
38 |
-
"total_flos":
|
39 |
-
"train_loss": 0.
|
40 |
-
"train_runtime":
|
41 |
-
"train_samples_per_second": 1.
|
42 |
-
"train_steps_per_second": 0.
|
43 |
}
|
44 |
],
|
45 |
"logging_steps": 500,
|
@@ -53,13 +53,13 @@
|
|
53 |
"should_epoch_stop": false,
|
54 |
"should_evaluate": false,
|
55 |
"should_log": false,
|
56 |
-
"should_save":
|
57 |
-
"should_training_stop":
|
58 |
},
|
59 |
"attributes": {}
|
60 |
}
|
61 |
},
|
62 |
-
"total_flos":
|
63 |
"train_batch_size": 2,
|
64 |
"trial_name": null,
|
65 |
"trial_params": null
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
+
"eval_loss": 0.7365190982818604,
|
14 |
+
"eval_runtime": 0.6967,
|
15 |
+
"eval_samples_per_second": 143.541,
|
16 |
+
"eval_steps_per_second": 18.66,
|
17 |
"step": 13
|
18 |
},
|
19 |
{
|
20 |
"epoch": 2.0,
|
21 |
+
"eval_loss": 0.9137925505638123,
|
22 |
+
"eval_runtime": 0.6958,
|
23 |
+
"eval_samples_per_second": 143.712,
|
24 |
+
"eval_steps_per_second": 18.683,
|
25 |
"step": 26
|
26 |
},
|
27 |
{
|
28 |
"epoch": 3.0,
|
29 |
+
"eval_loss": 0.9477354288101196,
|
30 |
+
"eval_runtime": 0.6964,
|
31 |
+
"eval_samples_per_second": 143.59,
|
32 |
+
"eval_steps_per_second": 18.667,
|
33 |
"step": 39
|
34 |
},
|
35 |
{
|
36 |
"epoch": 3.0,
|
37 |
"step": 39,
|
38 |
+
"total_flos": 648728369168384.0,
|
39 |
+
"train_loss": 0.4277030749198718,
|
40 |
+
"train_runtime": 188.9615,
|
41 |
+
"train_samples_per_second": 1.588,
|
42 |
+
"train_steps_per_second": 0.206
|
43 |
}
|
44 |
],
|
45 |
"logging_steps": 500,
|
|
|
53 |
"should_epoch_stop": false,
|
54 |
"should_evaluate": false,
|
55 |
"should_log": false,
|
56 |
+
"should_save": false,
|
57 |
+
"should_training_stop": false
|
58 |
},
|
59 |
"attributes": {}
|
60 |
}
|
61 |
},
|
62 |
+
"total_flos": 648728369168384.0,
|
63 |
"train_batch_size": 2,
|
64 |
"trial_name": null,
|
65 |
"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
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8bf45f26a4483f612d42291b1a443c8820a43ab24b747fd86c26449dada505f9
|
3 |
+
size 6520
|