Commits · inflaton-ai/logical-reasoning

try float32 for inference

d772d9d

dh-mc commited on Jul 21, 2024

best eval LF

0b5e165

inflaton commited on Jul 20, 2024

Merge branch 'main' of https://huggingface.co/spaces/smu-ai/logical-reasoning

e225773

HaotianHu commited on Jul 20, 2024

glm-4 p2 epoch 3

38d75e0

inflaton commited on Jul 20, 2024

Create tune-mgtv-qwen2_7b.sh

7c17cd1

HaotianHu commited on Jul 20, 2024

glm-4 p2 epoch 2 results

ae296ad

inflaton commited on Jul 20, 2024

glm-4 p1 results

1c71523

inflaton commited on Jul 20, 2024

glm-4 base p1 result

6599eac

inflaton commited on Jul 20, 2024

llama3 results

40ad559

inflaton commited on Jul 20, 2024

ready for tuning glm-4

29d9234

dh-mc commited on Jul 20, 2024

llama3 epoch 6 shows promising results; adding 4 epochs

16974cb

dh-mc commited on Jul 18, 2024

llama3 eval results

c4388ab

inflaton commited on Jul 18, 2024

merged epoch 1 of internlm2_5-7b-chat-1m_sft_bf16_p2_full

16ec7f8

inflaton commited on Jul 17, 2024

llama3 p2 results

3568465

inflaton commited on Jul 16, 2024

llama3 p2 epoch 1 results

4a18d7a

inflaton commited on Jul 16, 2024

new llama3 base results

4203a9b

inflaton commited on Jul 15, 2024

fixed bug for eval-mgtv-llama3_8b.sh

387046f

dh-mc commited on Jul 15, 2024

llama3 8b base model results

ad9400c

inflaton commited on Jul 15, 2024

switch llama3 version

c81ec3a

inflaton commited on Jul 15, 2024

working api/test

8ef46cb

inflaton commited on Jul 14, 2024

llama3

3a63ec4

dh-mc commited on Jul 14, 2024

attempted gemma2

bc0d2ff

inflaton commited on Jul 14, 2024

gemma2

0d2096e

dh-mc commited on Jul 14, 2024

Update tune-mgtv-qwen2_72b.sh

679dc6a

dh-mc commited on Jul 14, 2024

tune-mgtv-qwen2_72b.sh

09e6035

dh-mc commited on Jul 14, 2024

h100 results

df3521b

inflaton commited on Jul 14, 2024

sft_bf16_p2_full 4 epochs

f1e8d7b

inflaton commited on Jul 13, 2024

h100 4epochs for P1

93bc929

inflaton commited on Jul 13, 2024

updated script

645dc8d

inflaton commited on Jul 13, 2024

created scripts/tune-mgtv.sh

4e8cffe

dh-mc commited on Jul 13, 2024

llama-factory finetuning on Google Colab

36cb2cb

dh-mc commited on Jul 5, 2024

ref code from novel-translation

6702142

dh-mc commited on Jul 1, 2024

keep track of HT notebook

68936f4

dh-mc commited on Jun 28, 2024

minor bug fix

25aea89

dh-mc commited on Jun 27, 2024

initial code

6d1c39a

dh-mc commited on Jun 27, 2024

Spaces:

inflaton-ai
/

logical-reasoning

Build error

Commit History

try float32 for inference

d772d9d

best eval LF

0b5e165

Merge branch 'main' of https://huggingface.co/spaces/smu-ai/logical-reasoning

e225773

glm-4 p2 epoch 3

38d75e0

Create tune-mgtv-qwen2_7b.sh

7c17cd1

glm-4 p2 epoch 2 results

ae296ad

glm-4 p1 results

1c71523

glm-4 base p1 result

6599eac

llama3 results

40ad559

ready for tuning glm-4

29d9234

llama3 epoch 6 shows promising results; adding 4 epochs

16974cb

llama3 eval results

c4388ab

merged epoch 1 of internlm2_5-7b-chat-1m_sft_bf16_p2_full

16ec7f8

llama3 p2 results

3568465

llama3 p2 epoch 1 results

4a18d7a

new llama3 base results

4203a9b

fixed bug for eval-mgtv-llama3_8b.sh

387046f

llama3 8b base model results

ad9400c

switch llama3 version

c81ec3a

working api/test

8ef46cb

llama3

3a63ec4

attempted gemma2

bc0d2ff

gemma2

0d2096e

Update tune-mgtv-qwen2_72b.sh

679dc6a

tune-mgtv-qwen2_72b.sh

09e6035

h100 results

df3521b

sft_bf16_p2_full 4 epochs

f1e8d7b

h100 4epochs for P1

93bc929

updated script

645dc8d

created scripts/tune-mgtv.sh

4e8cffe

llama-factory finetuning on Google Colab

36cb2cb

ref code from novel-translation

6702142

keep track of HT notebook

68936f4

minor bug fix

25aea89

initial code

6d1c39a

Commit History

try float32 for inference d772d9d

best eval LF 0b5e165

Merge branch 'main' of https://huggingface.co/spaces/smu-ai/logical-reasoning e225773

glm-4 p2 epoch 3 38d75e0

Create tune-mgtv-qwen2_7b.sh 7c17cd1

glm-4 p2 epoch 2 results ae296ad

glm-4 p1 results 1c71523

glm-4 base p1 result 6599eac

llama3 results 40ad559

ready for tuning glm-4 29d9234

llama3 epoch 6 shows promising results; adding 4 epochs 16974cb

llama3 eval results c4388ab

merged epoch 1 of internlm2_5-7b-chat-1m_sft_bf16_p2_full 16ec7f8

llama3 p2 results 3568465

llama3 p2 epoch 1 results 4a18d7a

new llama3 base results 4203a9b

fixed bug for eval-mgtv-llama3_8b.sh 387046f

llama3 8b base model results ad9400c

switch llama3 version c81ec3a

working api/test 8ef46cb

llama3 3a63ec4

attempted gemma2 bc0d2ff

gemma2 0d2096e

Update tune-mgtv-qwen2_72b.sh 679dc6a

tune-mgtv-qwen2_72b.sh 09e6035

h100 results df3521b

sft_bf16_p2_full 4 epochs f1e8d7b

h100 4epochs for P1 93bc929

updated script 645dc8d

created scripts/tune-mgtv.sh 4e8cffe

llama-factory finetuning on Google Colab 36cb2cb

ref code from novel-translation 6702142

keep track of HT notebook 68936f4

minor bug fix 25aea89

initial code 6d1c39a

try float32 for inference

d772d9d

best eval LF

0b5e165

Merge branch 'main' of https://huggingface.co/spaces/smu-ai/logical-reasoning

e225773

glm-4 p2 epoch 3

38d75e0

Create tune-mgtv-qwen2_7b.sh

7c17cd1

glm-4 p2 epoch 2 results

ae296ad

glm-4 p1 results

1c71523

glm-4 base p1 result

6599eac

llama3 results

40ad559

ready for tuning glm-4

29d9234

llama3 epoch 6 shows promising results; adding 4 epochs

16974cb

llama3 eval results

c4388ab

merged epoch 1 of internlm2_5-7b-chat-1m_sft_bf16_p2_full

16ec7f8

llama3 p2 results

3568465

llama3 p2 epoch 1 results

4a18d7a

new llama3 base results

4203a9b

fixed bug for eval-mgtv-llama3_8b.sh

387046f

llama3 8b base model results

ad9400c

switch llama3 version

c81ec3a

working api/test

8ef46cb

llama3

3a63ec4

attempted gemma2

bc0d2ff

gemma2

0d2096e

Update tune-mgtv-qwen2_72b.sh

679dc6a

tune-mgtv-qwen2_72b.sh

09e6035

h100 results

df3521b

sft_bf16_p2_full 4 epochs

f1e8d7b

h100 4epochs for P1

93bc929

updated script

645dc8d

created scripts/tune-mgtv.sh

4e8cffe

llama-factory finetuning on Google Colab

36cb2cb

ref code from novel-translation

6702142

keep track of HT notebook

68936f4

minor bug fix

25aea89

initial code

6d1c39a