© 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. 10-RES_POST_32K-OAI
    11. 30025
    Prev
    Next
    INDEX
    Explanations

    phrases that involve analysis and comparison of data or statistics

    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
    131
    +0.11
    0.5%
    558
    +0.11
    0.5%
    300
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    455
    +0.11
    0.09
    469
    +0.11
    0.07
    654
    +0.10
    0.06
    Negative Logits
    bard
    -0.81
    ulhu
    -0.80
    xc
    -0.77
    natureconservancy
    -0.77
    camera
    -0.76
    ocracy
    -0.73
    sic
    -0.72
    /
    -0.68
    ocratic
    -0.68
    soever
    -0.67
    POSITIVE LOGITS
     averages
    1.09
     cumulative
    1.08
     percentages
    1.06
     median
    1.06
     total
    0.96
     trends
    0.94
     quarterly
    0.94
     respondents
    0.94
     totals
    0.93
     percentage
    0.92
    Activations Density 0.897%

    No Known Activations