© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 26982
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "among" and related phrases indicating groupings or associations

    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
    756
    +0.14
    0.7%
    112
    +0.09
    0.4%
    150
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    408
    +0.14
    0.11
    153
    +0.09
    0.07
    608
    +0.09
    0.10
    Negative Logits
    Basic
    -0.76
    bane
    -0.75
    ategory
    -0.75
    lvl
    -0.74
     Period
    -0.73
    ahime
    -0.72
    ensis
    -0.71
     Syndrome
    -0.70
    ldom
    -0.70
    dayName
    -0.69
    POSITIVE LOGITS
     recipients
    1.26
     beneficiaries
    1.10
     defendants
    1.09
     targets
    1.05
     speakers
    1.04
     participants
    1.03
     witnesses
    1.02
     accus
    1.02
     plaintiffs
    1.02
     victims
    1.01
    Activations Density 1.799%

    No Known Activations