Evaluations
Run models against your data
Introducing Evaluations, a powerful feature designed to enable you to effortlessly test and compare a selection of AI models against your datasets.
Whether you're fine-tuning models or evaluating performance metrics, Oxen evaluations simplifies the process, allowing you to quickly and easily run prompts through an entire dataset.
Once you're happy with the results, output the resulting dataset to a new file, another branch, or directly as a new commit.
ec1a052b-9429-44e8-81c8-c0f1c509fa93
OpenAIOpenAI/Dall-e 3textimage
Adam Singer
artforge
3 months ago
a simple svg line drawing of the outline of the profile view of a {name} airplane in black on a transparent background
completed 5 row sample0 tokens$ 0.0000 2 iterations
f5c764d6-e65e-4838-ad4d-2bb3a7f1aad6
OpenAIOpenAI/GPT-4otexttext
Adam Singer
artforge
3 months ago
in as few words as possible indicate the relationship of the sender of the sms message to the recipient
Text Message: {text}
Recipient: 
completed 10 row sample691 tokens$ 0.0022 24 iterations
1ffc25d1-05c3-4b72-9b47-7efb6cfa9a28
OpenAIOpenAI/GPT-4oimagetext
Adam Singer
artforge
3 months ago
write a brief (1 or 2 sentence) biography of the Simpsons character depicted in {image}
completed 11 rows8181 tokens$ 0.0244 3 iterations
8d0f8ea2-454e-4ebb-8bd5-3072866c638a
GoogleGoogle/gemini-1.5-proimagetext
Adam Singer
artforge
3 months ago
write a brief (1 or 2 sentence) biography of the Simpsons character depicted in {image}
completed 5 row sample0 tokens$ 0.0000 6 iterations
608da6ac-5886-443e-be55-b80c8ffbb839
OpenAIOpenAI/GPT-4oimagetext
Adam Singer
artforge
6 months ago
which character from the simpsons is shown in the image {image} respond with only the characters name and no additional explanation
description:
completed 11 rows7786 tokens$ 0.0198 9 iterations
333c3aaa-e4d7-4a86-adec-8dda8cb18cfd
OpenAIOpenAI/GPT-4o Minitexttext
Adam Singer
artforge
6 months ago
extract only the name of the character from the description
description: {description}
name: 
gpt4_name_branch
completed 11 rows450 tokens$ 0.0001 3 iterations
078bcbd0-b0cd-4860-b772-7dd9889cd3fe
OpenAIOpenAI/GPT-4o Minitexttext
Adam Singer
artforge
7 months ago
given iata aircraft code {iata_code} and aircraft name {name} return the name of the manuacturer of the aircraft with no additional text
completed 274 rows11907 tokens$ 0.0022 4 iterations
82c39d15-83f1-4500-a664-21bddcf42dc1
GoogleGoogle/Text Embedding 004textembeddings
Adam Singer
artforge
8 months ago
text
error An exception occurred indexing, getting dataframe and running evaluation: %Req.TransportError{reason: :closed} 40 / 100 rows0 tokens$ 0.0000 3 iterations
c61d7c35-d52d-4094-a40b-e8cfee146248
OpenAIOpenAI/Text Embeddings Small v3textembeddings
Adam Singer
artforge
8 months ago
text
error An exception occurred indexing, getting dataframe and running evaluation: %Req.TransportError{reason: :closed} 60 / 100 rows1586 tokens$ 0.0000 33 iterations
f1735c2a-4a84-41ec-8d7e-367d12d2f4a2
OpenAIOpenAI/Text Embeddings Small v3textembeddings
Adam Singer
artforge
8 months ago
text
started Running Waiting... 0 tokens$ 0.0000 1 iteration
fe75b3c0-2268-431f-a8c0-ddaf6f9d4e7b
OpenAIOpenAI/Text Embeddings Small v3textembeddings
Adam Singer
artforge
8 months ago
text
started Running Waiting... 0 tokens$ 0.0000 1 iteration
1430abf6-ee64-42d2-9554-3e66d2bbc24e
OpenAIOpenAI/GPT-4otextembeddings
Adam Singer
artforge
8 months ago
text
started Running Waiting... 0 tokens$ 0.0000 8 iterations
9be95567-3efd-474b-8177-98c688b72662
MetaMeta/Llama 3.1 405B Instruct Turbotexttext
Adam Singer
artforge
8 months ago
in as few words as possible indicate the relatonship of the sender of the sms message to the recipient
message: {text}
relationship:
completed 20 row sample2017 tokens$ 0.0071 31 iterations
c05e03bc-c334-435b-8dc8-6ca9ea238c72
OpenAIOpenAI/GPT-4o Minitexttext
Adam Singer
artforge
8 months ago
You are an expert system at reviewing SMS messages specializing in classifying the relationship of the sender to the recipient. in one word, indicate the relatonship of the sender of the sms message {text} to the recipient. respond only in one lowercased word
completed 100 rows8146 tokens$ 0.0049 1 iteration
445df47a-b580-4788-ac68-faf70961bb02
OpenAIOpenAI/GPT-4o Minitexttext
Adam Singer
artforge
8 months ago
in as few words as possible indicate the relatonship of the sender of the sms message {text} to the recipient
completed 5 row sample290 tokens$ 0.0002 2 iterations
1c936d4c-161c-4151-9541-5989b592e900
OpenAIOpenAI/GPT-4o Minitexttext
Adam Singer
artforge
8 months ago
in as few words as possible indicate the relatonship of the sender of the sms message {text} to the recipient
completed 100 rows5807 tokens$ 0.0035 2 iterations
cfb68d96-1ac3-4fac-a90b-c619d71aeb8c
OpenAIOpenAI/GPT-4otexttext
Adam Singer
artforge
9 months ago
in as few words as possible indicate the relatonship of the sender of the sms message {text} to the recipient
completed 100 rows5807 tokens 1 iteration
ae32d086-2b75-4c07-b5b4-d7524d4ffe68
OpenAIOpenAI/GPT-4otexttext
Adam Singer
artforge
9 months ago
in as few words as possible indicate the relatonship of the sender of the sms message {text} to the recipient
completed 100 rows5993 tokens 1 iteration