Edit model card

jinabert + gpt2 medium

This is a random-init encoder decoder from jinabert and GPT2. The max input length is 8192 and output length is 1024. Relevant tokenizers are in labeled branches.

Additionally, please note the bespoke software license (in addition to Apache 2.0):

"Commercial-Use Enforcer Apache 2.0 License"

**Preamble:**
This license governs the use of the accompanying software. The terms of the Apache License 2.0 fully apply, except as modified by the following additional clauses.

**1. Commercial Use Requirement:**
The software is provided under this license on the condition that it is ultimately used for commercial purposes. "Commercial purposes" refers to activities with the intent of monetary gain or other forms of business advantages.

**2. Proof of Commercial Use:**
a) Within one year of the initial use of the software, the user must provide proof of its commercial use. This proof can include, but is not limited to, financial records, business contracts, or other forms of evidence that clearly demonstrate the software’s use in a commercial context.

b) The proof must be mailed or faxed to the following coordinates: 48°52.6′S 123°23.6′W. The user is responsible for determining the appropriate method to send this proof to the specified coordinates.

**3. Consequences of Non-Compliance:**
Failure to provide proof of commercial use within the specified time frame will result in the immediate termination of the license. Upon termination, the user must cease all use of the software and destroy all copies, full or partial, of the software in their possession.

**4. General Legal Provisions:**
This license is subject to the general legal provisions outlined in the Apache License 2.0, including, but not limited to, limitations of liability, disclaimer of warranty, and copyright notice requirements.
Downloads last month
0
Safetensors
Model size
572M params
Tensor type
F32
·
Inference Examples
Inference API (serverless) has been turned off for this model.