© 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. 10-RES-JB
    11. 8520
    Prev
    Next
    INDEX
    Explanations

    the beginning of sentences, particularly clauses starting with "As"

    oai_token-act-pair · gpt-3.5-turbo

    phrases beginning with "As"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    thro
    -0.66
    abouts
    -0.65
    ALLY
    -0.63
    ãĤī
    -0.59
    ritic
    -0.59
    lean
    -0.58
     whine
    -0.58
     whatsoever
    -0.58
    âĻ¦
    -0.57
    didn
    -0.56
    POSITIVE LOGITS
    pects
    1.27
    semb
    1.22
    ymm
    1.15
    phalt
    1.13
    bestos
    1.13
    piring
    1.09
    ynchronous
    1.04
    semble
    1.02
    piration
    1.02
    king
    0.98
    Activations Density 0.066%

    No Known Activations