© 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. 8344
    Prev
    Next
    INDEX
    Explanations

    references to formal reports or studies related to social policies

    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
    424
    +0.11
    0.5%
    81
    +0.10
    0.4%
    176
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.11
    0.16
    385
    +0.10
    0.12
    393
    +0.09
    0.11
    Negative Logits
     chains
    -0.84
     verses
    -0.83
     headlines
    -0.81
     boots
    -0.78
     idols
    -0.76
     orbs
    -0.76
     mosques
    -0.75
     sacrifices
    -0.74
     favour
    -0.74
     oranges
    -0.74
    POSITIVE LOGITS
     Behavioral
    1.15
     Monitoring
    1.14
     Advisory
    1.10
     Regulatory
    1.05
     Disability
    1.02
     Prevention
    1.01
     Mobility
    1.01
     Balanced
    1.01
     Environmental
    1.01
     Domestic
    1.00
    Activations Density 1.183%

    No Known Activations