© 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 EvalsExportsAPI 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. 7025
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "it" and references to subjective experiences or feelings

    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
    410
    +0.12
    0.5%
    425
    +0.11
    0.5%
    656
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    566
    +0.12
    0.07
    139
    +0.11
    0.06
    477
    +0.10
    0.05
    Negative Logits
    conn
    -0.68
    cycle
    -0.67
     Hermes
    -0.66
    eers
    -0.66
     Scully
    -0.63
    \\\\\\\\
    -0.63
    ãĤ´
    -0.62
    RAY
    -0.61
     Grav
    -0.61
    hoff
    -0.61
    POSITIVE LOGITS
     warranted
    0.92
     slightest
    0.89
     least
    0.87
     widest
    0.84
     reasonably
    0.83
     disagreed
    0.79
     weakest
    0.79
     desired
    0.76
     comfortable
    0.76
     feas
    0.76
    Activations Density 0.626%

    No Known Activations