Model · gpt2-large
gpt2-large
Developer: OpenAI
Availability: available · checked 2026-09-24 · source
Raw record: /data/models/gpt2-large.json
Fields
- id
- gpt2-large
- identifiers
- huggingface
- openai-community/gpt2-large
- developer
- OpenAI
- release_date
- 2019-08-20partial · sourcenote: Follow-up post is itself the 774M release announcement.
- weights_status
- open
- availability
- available · checked 2026-09-24 · source
- license
- mitpartial · sourcenote: From HF card metadata (uploader-declared); confirm against the license text.
- architecture
- family
- decoder_only
- note
- family read from config 'architectures': ['GPT2LMHeadModel'].
- n_layers
- 36recorded · source
- hidden_size
- 1280recorded · source
- n_heads
- 20recorded · source
- vocab_size
- 50257recorded · source
- context_length
- 1024recorded · source
- positional_encoding
- learned_absoluterecorded · sourcenote: GPT-2 paper: model 'largely follows the details of the OpenAI GPT model'; GPT-1 paper: 'We used learned position embeddings instead of the sinusoidal version'.
- activation
- GELU (tanh approximation, config 'gelu_new')recorded · source
- training_data
- WebText: web pages linked from Reddit posts with karma >= 3. Same corpus as the 1.5B (XL) model.recorded · source
- techniques
- primary_sources
- record_history
- date:2026-09-24 · change:ingested as candidate from HF (openai-community/gpt2-large@32b71b12589c2f8d625668d2335a01cac3249519) · by:ingest_hf.py ·date:2026-09-24 · change:preparer: developer, release date, positional encoding, activation, trained_on set · by:claude (preparer, Sonnet 5) ·date:2026-09-24 · change:preparer cleanup: removed 2 stale flag(s), filled training_data · by:claude (preparer, Sonnet 5) ·date:2026-09-24 · change:reviewed and promoted from staging (1 edge(s) accepted) · by:Wilson Pruitt ·
Parents
Training data
Children
No edges recorded.
Read in
No station on the reading path has touched this record yet.