© 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. 10-RES_POST_32K-OAI
    11. 27891
    Prev
    Next
    INDEX
    Explanations

    references to proposals and plans related to projects or initiatives

    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
    764
    +0.12
    0.6%
    226
    +0.10
    0.4%
    72
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    86
    +0.12
    0.11
    55
    +0.10
    0.19
    72
    +0.09
    0.07
    Negative Logits
    ĸļ
    -0.96
     correction
    -0.79
     certs
    -0.76
     refunds
    -0.72
    icka
    -0.72
     Breach
    -0.72
     redress
    -0.71
     LIA
    -0.70
     Victim
    -0.70
     corros
    -0.70
    POSITIVE LOGITS
     brainstorm
    1.43
     envision
    1.40
     envis
    1.25
     contemplated
    1.24
     envisioned
    1.24
     proposal
    1.21
    bitious
    1.20
     floated
    1.17
     feasibility
    1.16
     proposals
    1.16
    Activations Density 12.414%

    No Known Activations