© 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. 8-RES-JB
    11. 17199
    Prev
    Next
    INDEX
    Explanations

    phrases expressing contrast or emphasis on a particular aspect

    oai_token-act-pair · gpt-3.5-turbo

    negative evaluations and criticisms of actions or situations

    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
     nonetheless
    -0.93
    etheless
    -0.73
     nevertheless
    -0.66
    cheat
    -0.63
    stress
    -0.62
     Finally
    -0.61
     Minotaur
    -0.61
     Lastly
    -0.61
    ague
    -0.58
    optional
    -0.58
    POSITIVE LOGITS
     aesthetics
    0.66
     pian
    0.59
     paycheck
    0.59
    ifiable
    0.59
    erves
    0.57
     coasts
    0.56
     aesthetic
    0.56
     but
    0.56
     nor
    0.55
     guiActive
    0.55
    Activations Density 0.204%

    No Known Activations