© 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. 618
    Prev
    Next
    INDEX
    Explanations

    statements or reports and categorize them

    oai_token-act-pair · gpt-3.5-turbo

    references to narratives or accounts involving events

    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
    omination
    -0.88
    omin
    -0.88
    sterdam
    -0.87
    inking
    -0.86
    ajor
    -0.85
    emale
    -0.84
    ategory
    -0.84
    ignt
    -0.82
    inence
    -0.82
    ateurs
    -0.77
    POSITIVE LOGITS
    telling
    0.97
     arc
    0.88
     revolving
    0.86
     arcs
    0.82
    story
    0.81
    tell
    0.79
    boards
    0.77
    board
    0.77
     stories
    0.77
     Stories
    0.76
    Activations Density 0.045%

    No Known Activations