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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 7329
    Prev
    Next
    INDEX
    Explanations

    mentions of the city of Paris, particularly in relation to climate agreements

    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
    438
    +0.10
    0.4%
    194
    +0.10
    0.4%
    461
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.10
    0.03
    511
    +0.10
    0.02
    145
    +0.09
    0.02
    Negative Logits
    ACTION
    -0.78
    ORD
    -0.67
    FER
    -0.65
     neck
    -0.64
    cloth
    -0.60
     darn
    -0.60
    FIELD
    -0.60
    ï¸ı
    -0.59
     judgement
    -0.59
    olds
    -0.59
    POSITIVE LOGITS
    ary
    1.22
    ism
    1.13
    alist
    1.07
    ists
    1.06
    ian
    1.06
    ity
    1.06
    ist
    1.05
    ité
    1.02
    istic
    1.02
    ate
    1.01
    Activations Density 0.114%

    No Known Activations