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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 5559
    Prev
    Next
    INDEX
    Explanations

    words related to undeclared or unspecified conditions, particularly in a political or economic context

    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
    28
    +0.12
    0.6%
    239
    +0.11
    0.5%
    120
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.12
    0.00
    232
    +0.11
    0.03
    659
    +0.11
    0.03
    Negative Logits
     goodwill
    -0.67
     OWN
    -0.61
     discounts
    -0.60
    oons
    -0.60
     precincts
    -0.60
    =-=-=-=-
    -0.60
     Haram
    -0.59
    rimp
    -0.58
     Nichols
    -0.56
    ishy
    -0.56
    POSITIVE LOGITS
    velop
    1.17
    ploy
    0.97
    ferred
    0.93
    bris
    0.93
    ceived
    0.93
    vent
    0.90
    lect
    0.89
    pend
    0.89
    herent
    0.88
    ven
    0.87
    Activations Density 0.171%

    No Known Activations