© 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. 8-RES_POST_32K-OAI
    6. 4402
    Prev
    Next
    INDEX
    Explanations

    the presence of parenthetical statements or citations in the text

    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
    425
    +0.11
    0.5%
    291
    +0.11
    0.5%
    220
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.11
    0.11
    314
    +0.11
    0.03
    627
    +0.10
    0.05
    Negative Logits
    rawdownloadcloneembedreportprint
    -0.76
     Intake
    -0.75
     Distribution
    -0.71
    Reviewer
    -0.71
     lag
    -0.68
     nic
    -0.64
     variance
    -0.62
    ####
    -0.62
     lvl
    -0.60
    Reply
    -0.60
    POSITIVE LOGITS
    along
    1.19
    albeit
    1.19
    successfully
    1.07
    among
    1.04
    often
    1.02
    though
    1.00
    pictured
    0.98
    literally
    0.97
    including
    0.97
    particularly
    0.96
    Activations Density 0.296%

    No Known Activations