© 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. 4-RES-JB
    11. 16767
    Prev
    Next
    INDEX
    Explanations

    phrases or words related to things that are not packaged or compressed

    oai_token-act-pair · gpt-3.5-turbo

    words that refer to something being unpacked or uncompromising

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

    No Comments

    Negative Logits
    姫
    -0.90
     Lauder
    -0.82
     Carbuncle
    -0.82
    anwhile
    -0.80
    quickShipAvailable
    -0.79
     Rams
    -0.76
    */(
    -0.72
    tone
    -0.69
    Interstitial
    -0.68
     obser
    -0.67
    POSITIVE LOGITS
     unp
    1.16
    orter
    1.14
    aired
    1.10
    ublic
    1.06
    ainted
    1.05
    ardon
    0.99
    icked
    0.99
    ressed
    0.99
    icking
    0.97
    iqu
    0.95
    Activations Density 0.011%

    No Known Activations