© 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 EvalsExportsinterp-engineNEWAPI 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. 7882
    Prev
    Next
    INDEX
    Explanations

    phrases emphasizing the relationship between actions and their consequences

    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
    627
    +0.12
    0.6%
    320
    +0.09
    0.4%
    476
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    88
    +0.12
    0.06
    108
    +0.09
    0.07
    608
    +0.09
    0.09
    Negative Logits
    advertisement
    -0.72
     Unsure
    -0.69
    Vers
    -0.68
     showc
    -0.67
    Refer
    -0.67
    worn
    -0.65
    etting
    -0.63
    Upgrade
    -0.62
    VERTISEMENT
    -0.61
    FW
    -0.61
    POSITIVE LOGITS
     anybody
    1.16
     everybody
    1.15
     humankind
    1.12
     whoever
    1.11
    everyone
    1.09
     anyone
    1.08
     everyone
    1.06
     mankind
    1.06
     ours
    0.99
     yours
    0.99
    Activations Density 1.067%

    No Known Activations