© 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. 5705
    Prev
    Next
    INDEX
    Explanations

    references to depth in various contexts

    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
    99
    +0.16
    0.7%
    195
    +0.15
    0.7%
    549
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.16
    0.02
    195
    +0.15
    0.01
    194
    +0.09
    0.01
    Negative Logits
    eer
    -1.00
    eers
    -0.97
    jee
    -0.81
    eering
    -0.76
     Sabha
    -0.69
    akedown
    -0.66
    immune
    -0.65
    RY
    -0.64
    abad
    -0.64
    RON
    -0.62
    POSITIVE LOGITS
     perception
    0.89
     depth
    0.78
    chart
    0.71
    plings
    0.71
    hole
    0.70
     penetration
    0.69
     dug
    0.68
     Depth
    0.67
    cre
    0.66
     depths
    0.66
    Activations Density 0.041%

    No Known Activations