Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 6-RES_POST_32K-OAI
    6. 4238
    Prev
    Next
    INDEX
    Explanations

    contrasting conjunctions that imply a shift in reasoning or perspective

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_6.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.6.hook_resid_post
    Hook Layer
    6
    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
    71
    +0.12
    0.5%
    336
    +0.09
    0.4%
    191
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    572
    +0.12
    0.04
    135
    +0.09
    0.04
    8
    +0.09
    0.04
    Negative Logits
    ggles
    -0.80
    atars
    -0.75
    ventions
    -0.74
    odcast
    -0.73
    idences
    -0.71
    uthor
    -0.71
    character
    -0.68
    controlled
    -0.68
    ident
    -0.68
    regular
    -0.67
    POSITIVE LOGITS
     cd
    0.70
    Merit
    0.68
     nor
    0.67
     consolation
    0.61
     insofar
    0.60
     Mehran
    0.59
    amaz
    0.59
    atche
    0.57
    .>>
    0.57
     rhet
    0.57
    Activations Density 0.170%

    No Known Activations