© 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. 5700
    Prev
    Next
    INDEX
    Explanations

    phrases related to interruptions or pauses in work activities

    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
    601
    +0.13
    0.6%
    373
    +0.11
    0.5%
    497
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    28
    +0.13
    0.03
    728
    +0.11
    0.02
    744
    +0.11
    0.03
    Negative Logits
    worn
    -0.70
    swick
    -0.69
     jewels
    -0.67
    nai
    -0.67
    selves
    -0.67
    redients
    -0.64
     formulas
    -0.64
     sophistication
    -0.64
    igree
    -0.63
    gewater
    -0.63
    POSITIVE LOGITS
    motion
    0.78
    ão
    0.77
    ©¶æ
    0.75
     ado
    0.71
    ŃĶ
    0.71
     impede
    0.70
    fm
    0.68
     disbelief
    0.68
    assment
    0.67
    endment
    0.66
    Activations Density 0.442%

    No Known Activations