© 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. 1-RES-JB
    11. 8987
    Prev
    Next
    INDEX
    Explanations

    the word 'hard' as part of a token in the text

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "hard."

    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
    uality
    -0.73
    ript
    -0.66
     tidy
    -0.64
     Sut
    -0.63
     Nile
    -0.62
     Kislyak
    -0.62
     Spect
    -0.61
     chart
    -0.60
     Neph
    -0.59
     Waterloo
    -0.58
    POSITIVE LOGITS
    hard
    1.23
    ware
    1.06
    ened
    1.02
    ball
    0.92
    hardt
    0.90
    ening
    0.89
    rod
    0.86
    castle
    0.86
    balls
    0.84
    icult
    0.83
    Activations Density 0.005%

    No Known Activations