© 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. 10-RES_POST_32K-OAI
    11. 29445
    Prev
    Next
    INDEX
    Explanations

    instances of the word "fl" which indicates a focus on slang or informal expressions

    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
    289
    +0.12
    0.5%
    712
    +0.10
    0.4%
    233
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    103
    +0.12
    0.03
    232
    +0.10
    0.03
    289
    +0.10
    0.01
    Negative Logits
     caster
    -0.69
     Claus
    -0.68
     sorcerer
    -0.67
    byn
    -0.64
     Claire
    -0.62
    renheit
    -0.60
     Miko
    -0.59
    pole
    -0.59
     peace
    -0.58
    cius
    -0.58
    POSITIVE LOGITS
    aunted
    1.07
    aunting
    0.91
    ust
    0.90
    outed
    0.87
    outing
    0.87
    semb
    0.84
    inging
    0.84
    IPP
    0.83
    irted
    0.82
    ipel
    0.80
    Activations Density 0.214%

    No Known Activations