© 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
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 7021
    Prev
    Next
    INDEX
    Explanations

    colons and their associated usage in punctuation and formatting

    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
    349
    +0.10
    0.5%
    373
    +0.10
    0.5%
    136
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    394
    +0.10
    0.01
    440
    +0.10
    0.01
    683
    +0.10
    0.01
    Negative Logits
     thereto
    -0.69
    alam
    -0.69
     <-
    -0.68
    Els
    -0.62
    versions
    -0.62
    anwhile
    -0.62
    wcs
    -0.61
    â̦â̦â̦â̦â̦â̦â̦â̦
    -0.61
    etc
    -0.61
    bery
    -0.60
    POSITIVE LOGITS
    acan
    0.70
    ceived
    0.63
    iuses
    0.62
    vin
    0.61
    roach
    0.60
    otta
    0.60
    amera
    0.59
    inion
    0.59
    overe
    0.58
    essim
    0.57
    Activations Density 0.041%

    No Known Activations