© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 31510
    Prev
    Next
    INDEX
    Explanations

    numerical values or calculations

    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
    620
    +0.14
    0.6%
    311
    +0.13
    0.6%
    145
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    442
    +0.14
    0.08
    620
    +0.13
    0.04
    699
    +0.10
    0.06
    Negative Logits
     severed
    -0.60
     oath
    -0.59
    ability
    -0.58
    erity
    -0.57
    ercise
    -0.57
     irresist
    -0.56
    krit
    -0.56
     conduc
    -0.56
    lehem
    -0.55
    perty
    -0.55
    POSITIVE LOGITS
    xff
    1.42
    xa
    1.12
    xd
    1.10
    xb
    1.06
    x
    1.04
    xc
    1.01
    xe
    1.01
    603
    0.90
    644
    0.90
    755
    0.89
    Activations Density 0.600%

    No Known Activations