© 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. 11-RES-JB
    11. 5120
    Prev
    Next
    INDEX
    Explanations

    phrases related to the absence of control or limit

    oai_token-act-pair · gpt-3.5-turbo

    affirmative statements about the qualities or states of entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    inav
    -0.97
    ãĤ©
    -0.83
     Moines
    -0.75
     Turns
    -0.74
    iets
    -0.72
    aptic
    -0.72
    iasm
    -0.72
    vous
    -0.72
    icago
    -0.70
    ffe
    -0.70
    POSITIVE LOGITS
     unfit
    1.31
     unsafe
    1.23
     unsu
    1.20
     unworthy
    1.13
     worthless
    1.12
     inherently
    1.10
     unprotected
    1.07
     incompatible
    1.07
     vulnerable
    1.05
     sacred
    1.04
    Activations Density 0.252%

    No Known Activations