© 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. 8022
    Prev
    Next
    INDEX
    Explanations

    verbs related to actions indicating reversal or undoing

    oai_token-act-pair · gpt-3.5-turbo

    actions related to reversing or undoing policies

    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
    rics
    -0.82
    bags
    -0.75
    chens
    -0.74
    raq
    -0.71
    azines
    -0.68
    munition
    -0.65
    rug
    -0.63
    mpeg
    -0.63
    riz
    -0.62
    rosso
    -0.62
    POSITIVE LOGITS
     havoc
    0.76
    laus
    0.72
    ettings
    0.72
     decades
    0.71
     Genie
    0.70
    graded
    0.66
     Roe
    0.66
    CLASSIFIED
    0.65
     longstanding
    0.65
     course
    0.65
    Activations Density 0.088%

    No Known Activations