© 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. 10-RES_POST_32K-OAI
    11. 29526
    Prev
    Next
    INDEX
    Explanations

    phrases related to claims, skepticism, and the refutation of arguments or misinformation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    +0.11
    0.5%
    114
    +0.10
    0.4%
    291
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    386
    +0.11
    0.10
    679
    +0.10
    0.12
    401
    +0.10
    0.10
    Negative Logits
     actionGroup
    -0.81
     Lumpur
    -0.80
     dues
    -0.79
     horizont
    -0.71
    rontal
    -0.70
    gencies
    -0.70
    ktop
    -0.69
     happier
    -0.69
     downtime
    -0.69
     Compact
    -0.68
    POSITIVE LOGITS
     debunked
    1.43
     refuted
    1.19
     baseless
    1.13
    False
    1.12
     falsely
    1.10
    Claim
    1.09
     debunk
    1.09
     discredited
    1.03
     assertions
    1.02
     asserted
    1.00
    Activations Density 1.571%

    No Known Activations