© 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. 2165
    Prev
    Next
    INDEX
    Explanations

    the word "cub" at differing levels of specificity and in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    references to 'cubs' or young animals, particularly related to tigers or bears

    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
    wine
    -0.86
    ENGTH
    -0.67
    backer
    -0.67
     Lumpur
    -0.67
    Downloadha
    -0.66
     IST
    -0.64
     irritating
    -0.64
    tenance
    -0.63
    ARP
    -0.61
    EEP
    -0.60
    POSITIVE LOGITS
    ebook
    1.06
    elight
    1.04
    icles
    1.02
    icle
    1.00
    eless
    0.96
    ility
    0.94
    MpServer
    0.90
    bies
    0.90
    zees
    0.87
    rium
    0.86
    Activations Density 0.099%

    No Known Activations