© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 3-RES-JB
    11. 5320
    Prev
    Next
    INDEX
    Explanations

    comparative words and phrases, particularly focusing on the concept of "bigger."

    oai_token-act-pair · gpt-3.5-turbo

    references to "bigger" concepts or items in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    shire
    -0.91
    ARA
    -0.74
    endants
    -0.73
    endant
    -0.70
    EVA
    -0.69
    Grade
    -0.68
    ibel
    -0.68
    pta
    -0.67
    syn
    -0.67
    onso
    -0.67
    POSITIVE LOGITS
     than
    1.03
     Than
    0.91
     eleph
    0.87
    than
    0.84
    izont
    0.82
     sized
    0.80
    oted
    0.76
     fish
    0.74
     hippocamp
    0.74
     bang
    0.74
    Activations Density 0.010%

    No Known Activations