© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 29896
    Prev
    Next
    INDEX
    Explanations

    occurrences of the phrase "in" indicating context or relevance in discussions

    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
    608
    +0.14
    0.6%
    669
    +0.10
    0.5%
    715
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.14
    0.07
    669
    +0.10
    0.06
    641
    +0.10
    0.05
    Negative Logits
    lag
    -0.79
    reau
    -0.76
    Fail
    -0.75
    tale
    -0.68
    rical
    -0.67
    orst
    -0.66
    soever
    -0.65
    NESS
    -0.65
    rete
    -0.65
    cue
    -0.65
    POSITIVE LOGITS
     seeing
    1.16
     pursuing
    1.11
     hearing
    1.08
     exploring
    1.07
     acquiring
    1.05
     learning
    1.02
     preserving
    0.99
     studying
    0.97
     knowing
    0.97
     joining
    0.96
    Activations Density 0.362%

    No Known Activations