© 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 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. 30463
    Prev
    Next
    INDEX
    Explanations

    references to comparisons or contrasts in different contexts

    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
    560
    +0.11
    0.5%
    95
    +0.10
    0.4%
    597
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    479
    +0.11
    0.02
    669
    +0.10
    0.09
    6
    +0.10
    0.11
    Negative Logits
    arettes
    -0.80
    phys
    -0.76
    STD
    -0.74
    osterone
    -0.74
    wire
    -0.70
    tails
    -0.68
    arella
    -0.68
     surg
    -0.67
    Eth
    -0.67
    ebin
    -0.66
    POSITIVE LOGITS
     Great
    2.10
    great
    2.06
    Great
    2.04
     GREAT
    1.85
     great
    1.80
    grand
    1.57
     grand
    1.50
     Grand
    1.49
     Greatest
    1.42
    Grand
    1.41
    Activations Density 3.567%

    No Known Activations