© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 11-RES-JB
    11. 11754
    Prev
    Next
    INDEX
    Explanations

    phrases comparing different entities or concepts using the word "that" and drawing parallels between them

    oai_token-act-pair · gpt-3.5-turbo

    comparative phrases emphasizing 'that' in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    APTER
    -0.64
    glers
    -0.63
    orously
    -0.61
    Numbers
    -0.61
    ãĥĥ
    -0.60
    zos
    -0.60
    aths
    -0.58
    Textures
    -0.56
    aji
    -0.56
    Leaks
    -0.54
    POSITIVE LOGITS
     of
    1.22
     afforded
    1.05
     seen
    0.96
     enjoyed
    0.96
     pioneered
    0.93
     produced
    0.93
     exhibited
    0.92
     undertaken
    0.91
     pertaining
    0.91
     generated
    0.90
    Activations Density 0.071%

    No Known Activations