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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 11427
    Prev
    Next
    INDEX
    Explanations

    special characters like quotes («, »), arrows (-->), and symbols (âĢº)

    oai_token-act-pair · gpt-3.5-turbo

    special characters or symbols used in formatting or punctuation

    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
    )=(
    -0.68
    uint
    -0.55
     dimensional
    -0.53
    [
    -0.52
     suppressing
    -0.52
     discrim
    -0.51
    msg
    -0.51
    ochem
    -0.51
     Euras
    -0.50
    Sham
    -0.50
    POSITIVE LOGITS
     »
    4.28
     âĢº
    1.70
     «
    1.70
    »
    1.56
     -->
    1.45
     >>
    1.43
    «
    1.32
     âĨĴ
    1.25
     âĸ
    1.24
     ·
    1.23
    Activations Density 0.015%

    No Known Activations