© 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. 4121
    Prev
    Next
    INDEX
    Explanations

    phrases related to possession or ownership of items

    oai_token-act-pair · gpt-3.5-turbo

    the word "some" in various contexts throughout the document

    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
    SPONSORED
    -0.89
    enance
    -0.89
    agents
    -0.78
    raid
    -0.74
    gon
    -0.74
    ettes
    -0.70
    ean
    -0.68
    letters
    -0.68
    sburgh
    -0.68
     borne
    -0.67
    POSITIVE LOGITS
    place
    1.15
    ones
    0.99
    how
    0.89
    body
    0.88
     semblance
    0.86
     degree
    0.81
     consolation
    0.80
     sort
    0.75
     mornings
    0.75
     additional
    0.73
    Activations Density 0.111%

    No Known Activations