© 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. 1-RES-JB
    11. 13237
    Prev
    Next
    INDEX
    Explanations

    functions or features that allow a user to modify or customize something

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate functionality or capabilities of an application or system

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Correspond
    -0.70
    Ô
    -0.69
     Sins
    -0.67
    ?????-
    -0.66
     Commonwealth
    -0.64
     Biblical
    -0.62
     Samoa
    -0.62
     Beasts
    -0.61
     Galile
    -0.60
     Abyssal
    -0.60
    POSITIVE LOGITS
    heet
    1.04
    terness
    0.93
    tered
    0.88
    sheet
    0.85
     us
    0.84
    terday
    0.81
    tering
    0.79
    creen
    0.79
    ettings
    0.78
    eries
    0.78
    Activations Density 0.005%

    No Known Activations