© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 23558
    Prev
    Next
    INDEX
    Explanations

    images with the word "Baby" in the description followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    terms related to baby animals or photography, alongside references to specific weeks or phases in a schedule

    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
     ..."
    -0.75
     ...
    -0.65
     \"
    -0.63
     transf
    -0.60
    ..."
    -0.59
    .</
    -0.57
     Morty
    -0.57
     ''
    -0.56
    ,...
    -0.56
     EVs
    -0.54
    POSITIVE LOGITS
    intosh
    1.01
    withstanding
    0.94
    resa
    0.92
    jamin
    0.91
    foundland
    0.90
    etheless
    0.90
    agascar
    0.86
    theless
    0.84
    odore
    0.83
    gotten
    0.83
    Activations Density 0.434%

    No Known Activations