© 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. 4586
    Prev
    Next
    INDEX
    Explanations

    words associated with actions and their impacts

    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
    218
    +0.17
    0.8%
    526
    +0.13
    0.6%
    203
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.17
    0.01
    229
    +0.13
    0.01
    63
    +0.12
    0.00
    Negative Logits
    icht
    -0.68
     upkeep
    -0.63
     hitch
    -0.63
    rouch
    -0.63
    ointed
    -0.62
     tatt
    -0.61
     Scotia
    -0.60
    inances
    -0.60
    rehend
    -0.58
    illon
    -0.58
    POSITIVE LOGITS
    ãĥ³
    1.33
    ãĤ¤
    1.27
    ãĥĦ
    1.15
    ãĥĥ
    1.11
    ãĤ®
    1.06
    ãĤ¹
    1.00
    ãĥİ
    1.00
    ãĤ¤ãĥĪ
    0.99
    ãĥª
    0.97
    ãĤ·
    0.97
    Activations Density 0.010%

    No Known Activations