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

    Neuronpedia

    Jacobian LensNEW
    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. 8-RES-JB
    11. 11135
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a strong disclaimer or difference of opinion

    oai_token-act-pair · gpt-3.5-turbo

    phrases that assert the non-existence or non-affiliation of something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     glim
    -0.69
    itone
    -0.65
    anmar
    -0.62
    essor
    -0.61
    idges
    -0.60
    ilts
    -0.60
    oute
    -0.60
    eport
    -0.58
    iture
    -0.58
    erva
    -0.58
    POSITIVE LOGITS
     whatsoever
    1.29
     resembling
    0.93
     resembles
    0.93
     resembled
    0.87
     harmed
    0.80
     resemble
    0.79
     connected
    0.77
     detract
    0.76
    soever
    0.72
    Reply
    0.70
    Activations Density 0.053%

    No Known Activations