© 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. 0-RES-JB
    11. 8406
    Prev
    Next
    INDEX
    Explanations

    phrases related to upholstery

    oai_token-act-pair · gpt-3.5-turbo

    instances of the substring "hol" in words

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Carbuncle
    -0.75
     Mega
    -0.69
     Node
    -0.65
     Isis
    -0.64
     Pyrrha
    -0.59
     unexpectedly
    -0.59
     identical
    -0.59
     Antonio
    -0.59
     parity
    -0.58
     formation
    -0.57
    POSITIVE LOGITS
    hol
    4.86
    hole
    1.64
    holes
    1.60
    Hol
    1.59
     hol
    1.47
    hal
    1.36
     Hol
    1.35
    hart
    1.28
    hov
    1.25
     HOL
    1.24
    Activations Density 0.003%

    No Known Activations