© 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. 8-RES_POST_32K-OAI
    11. 5633
    Prev
    Next
    INDEX
    Explanations

    words related to medical conditions and their descriptions

    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
    343
    +0.12
    0.6%
    36
    +0.11
    0.5%
    372
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.12
    0.12
    343
    +0.11
    0.09
    406
    +0.11
    0.08
    Negative Logits
    sed
    -0.80
    ty
    -0.78
    t
    -0.77
    deck
    -0.74
    pick
    -0.73
    sin
    -0.72
    ties
    -0.71
    lasses
    -0.71
    n
    -0.68
    mare
    -0.68
    POSITIVE LOGITS
    uthor
    1.40
    ñ
    1.38
    BILITY
    1.23
    ÅŁ
    1.20
    issance
    1.16
    ï
    1.11
    vel
    1.10
    ð
    1.07
    BILITIES
    1.07
    qt
    1.03
    Activations Density 1.036%

    No Known Activations