© 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. 5947
    Prev
    Next
    INDEX
    Explanations

    negative numerical values or indicators

    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
    373
    +0.16
    0.7%
    640
    +0.13
    0.6%
    314
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    71
    +0.16
    0.02
    640
    +0.13
    0.02
    739
    +0.11
    0.01
    Negative Logits
     decon
    -0.68
     bin
    -0.65
    wagen
    -0.62
     task
    -0.62
     actionGroup
    -0.60
     belie
    -0.60
     imagining
    -0.60
     furthe
    -0.59
     sacrific
    -0.58
    itability
    -0.57
    POSITIVE LOGITS
     Advertisement
    1.21
    Fill
    0.92
    =-=-=-=-
    0.91
     Stuff
    0.83
    End
    0.83
    Cross
    0.80
    Commercial
    0.79
    Names
    0.79
    Population
    0.78
    ¯¯¯¯¯¯¯¯
    0.76
    Activations Density 0.052%

    No Known Activations