© 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. 24282
    Prev
    Next
    INDEX
    Explanations

    mentions of specific names or terms, such as "Bungalow Hendrick" and "drunken."

    oai_token-act-pair · gpt-3.5-turbo

    words associated with drunkenness and specific individuals or entities

    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
    terday
    -0.78
    ocene
    -0.76
    ipop
    -0.74
    ovember
    -0.72
     Classification
    -0.66
    ptions
    -0.65
    代
    -0.65
     ILCS
    -0.65
     PRESS
    -0.65
     Passage
    -0.65
    POSITIVE LOGITS
    liness
    1.03
    led
    0.99
    ling
    0.96
    lesh
    0.94
    lins
    0.92
    rah
    0.88
    gob
    0.86
    redo
    0.86
    orough
    0.85
    leneck
    0.85
    Activations Density 0.028%

    No Known Activations