© 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. 0-RES-JB
    11. 18879
    Prev
    Next
    INDEX
    Explanations

    phrases mentioning or related to someone named "Bill Sharp"

    oai_token-act-pair · gpt-3.5-turbo

    repeated instances of the letters "SH" in a sequence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    gio
    -0.75
    eur
    -0.70
     Afric
    -0.66
    ious
    -0.65
    iously
    -0.65
    antine
    -0.63
    ophon
    -0.63
    Äĩ
    -0.62
     nomine
    -0.62
    iere
    -0.62
    POSITIVE LOGITS
     SH
    3.56
    SH
    2.27
     SHAR
    1.55
     Sh
    1.44
    Sh
    1.42
    sh
    1.36
     SAL
    1.27
     SHE
    1.25
     SL
    1.25
     SAM
    1.24
    Activations Density 0.012%

    No Known Activations