© 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. 1-RES-JB
    11. 10233
    Prev
    Next
    INDEX
    Explanations

    words indicating strong emphasis or negation, such as 'EVERYTHING', 'VERY', 'NOT', 'NEVER', 'ALWAYS'

    oai_token-act-pair · gpt-3.5-turbo

    emphatic negations and the word "every."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    idem
    -0.80
     Slovenia
    -0.75
    ipel
    -0.73
    atz
    -0.73
    ijk
    -0.72
    opol
    -0.72
    orio
    -0.71
     Cheong
    -0.70
    gio
    -0.69
    idable
    -0.68
    POSITIVE LOGITS
    THING
    1.32
     MUCH
    1.21
     NOT
    1.17
     FUCK
    1.16
     THERE
    1.14
     WITH
    1.14
    ELY
    1.13
    ONE
    1.13
     WAY
    1.10
     THEN
    1.10
    Activations Density 0.051%

    No Known Activations