© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 9462
    Prev
    Next
    INDEX
    Explanations

    measurements of distance in kilometers

    oai_token-act-pair · gpt-3.5-turbo

    opening parentheses in numerical contexts

    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
     lax
    -0.79
     lull
    -0.70
     reneg
    -0.68
     Phar
    -0.66
     fle
    -0.66
    quished
    -0.63
     handlers
    -0.62
     wre
    -0.62
     dro
    -0.62
     spot
    -0.61
    POSITIVE LOGITS
    sic
    1.21
    excluding
    1.21
    including
    1.19
    ...)
    1.19
    â̦)
    1.17
    formerly
    1.12
    approximately
    1.12
    !)
    1.11
    depending
    1.11
    possibly
    1.10
    Activations Density 0.108%

    No Known Activations