© 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. 4531
    Prev
    Next
    INDEX
    Explanations

    instances of the word "that" and its variations within various contexts

    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
    378
    +0.10
    0.5%
    420
    +0.10
    0.4%
    436
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    161
    +0.10
    0.03
    662
    +0.10
    0.03
    229
    +0.08
    0.03
    Negative Logits
    theless
    -0.64
     reconciliation
    -0.61
    bit
    -0.55
     suspicion
    -0.54
     defiance
    -0.53
     micro
    -0.53
     onward
    -0.53
     mediation
    -0.53
    sheet
    -0.53
     quiet
    -0.53
    POSITIVE LOGITS
     seen
    0.96
     presented
    0.95
     portrayed
    0.93
     depicted
    0.93
     afforded
    0.91
     encountered
    0.91
     shown
    0.89
     utilized
    0.89
     enjoyed
    0.89
     bestowed
    0.88
    Activations Density 0.162%

    No Known Activations