© 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
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 10-RES_POST_32K-OAI
    6. 29470
    Prev
    Next
    INDEX
    Explanations

    mentions of injuries in various contexts

    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
    151
    +0.11
    0.5%
    258
    +0.10
    0.5%
    651
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    759
    +0.11
    0.04
    651
    +0.10
    0.04
    189
    +0.10
    0.04
    Negative Logits
    £ı
    -0.72
    riber
    -0.72
    hyde
    -0.71
    xual
    -0.69
    NRS
    -0.69
    DN
    -0.68
    chin
    -0.67
    rers
    -0.67
    imeters
    -0.66
    ocom
    -0.66
    POSITIVE LOGITS
     plagued
    0.93
     woes
    0.89
     plag
    0.79
     bug
    0.77
     sidelined
    0.76
     scare
    0.76
     ravaged
    0.75
     troubles
    0.75
     setback
    0.72
     bothering
    0.71
    Activations Density 0.602%

    No Known Activations