© 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 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. 28187
    Prev
    Next
    INDEX
    Explanations

    financial figures and statistics within a context related to political or economic events

    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
    455
    +0.13
    0.6%
    525
    +0.10
    0.4%
    438
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    530
    +0.13
    0.06
    6
    +0.10
    0.07
    430
    +0.09
    0.05
    Negative Logits
    Ps
    -1.16
    Anim
    -1.13
    Leaks
    -1.07
    Enh
    -1.06
    Person
    -1.05
    Style
    -1.02
    Stream
    -1.01
    Planet
    -1.00
    Problem
    -0.99
    Import
    -0.99
    POSITIVE LOGITS
    leon
    0.89
     thirds
    0.79
     baker
    0.75
     msec
    0.70
     SAR
    0.70
     frank
    0.69
     lettuce
    0.69
    hyde
    0.69
     mc
    0.68
     KR
    0.68
    Activations Density 0.316%

    No Known Activations