© 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. 23079
    Prev
    Next
    INDEX
    Explanations

    instances of the word or suffix 'fter' followed by a numeric activation value, particularly when the activation value is higher

    oai_token-act-pair · gpt-3.5-turbo

    instances of the suffix "fter" and words related to publishing

    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
     precincts
    -0.67
     GOODMAN
    -0.63
     subordinates
    -0.63
     presidents
    -0.62
    oph
    -0.61
     policing
    -0.60
     siph
    -0.59
     tanks
    -0.59
     governors
    -0.59
     roam
    -0.58
    POSITIVE LOGITS
    fter
    1.21
    lishes
    1.02
    lish
    0.94
    fters
    0.91
    bler
    0.90
    Lago
    0.84
    rue
    0.83
    mens
    0.78
    forth
    0.76
    SourceFile
    0.76
    Activations Density 0.009%

    No Known Activations