© 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
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 6041
    Prev
    Next
    INDEX
    Explanations

    causal conjunctions and conditional phrases

    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
    223
    +0.12
    0.6%
    715
    +0.12
    0.5%
    50
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.12
    0.05
    655
    +0.12
    0.04
    647
    +0.12
    0.04
    Negative Logits
    itiz
    -1.00
    enium
    -0.70
    hz
    -0.70
    tty
    -0.69
    FML
    -0.68
    wcs
    -0.66
    HUD
    -0.65
    emi
    -0.65
    istries
    -0.63
    WN
    -0.63
    POSITIVE LOGITS
     simultane
    0.61
     congratulations
    0.60
     congr
    0.59
     apologies
    0.59
     presumably
    0.59
     moreover
    0.58
     encyclopedia
    0.57
     um
    0.56
     perhaps
    0.55
     irony
    0.55
    Activations Density 0.308%

    No Known Activations