© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 4793
    Prev
    Next
    INDEX
    Explanations

    names of authors and researchers associated with academic publications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    373
    +0.23
    1.1%
    38
    +0.10
    0.5%
    495
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.23
    0.13
    232
    +0.10
    0.11
    373
    +0.10
    -0.03
    Negative Logits
     Vaughn
    -0.68
     Bucket
    -0.67
     Coliseum
    -0.67
     Millenn
    -0.66
     Shakespeare
    -0.64
     Cedar
    -0.62
     Cain
    -0.60
     Hancock
    -0.60
     segregated
    -0.60
     handshake
    -0.60
    POSITIVE LOGITS
    âĢIJ
    0.97
    ÄŁ
    0.95
    ón
    0.90
    ás
    0.90
    Ã
    0.89
    å
    0.89
    ö
    0.88
    ÅĤ
    0.87
    ão
    0.85
    oslav
    0.84
    Activations Density 0.532%

    No Known Activations