© 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. 3349
    Prev
    Next
    INDEX
    Explanations

    technical descriptions or comparisons between various items

    oai_token-act-pair · gpt-3.5-turbo

    relative clauses and references that provide additional information about subjects or objects

    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
     Kazakh
    -0.66
     Skydragon
    -0.62
    SE
    -0.62
    å½
    -0.62
    Bone
    -0.62
     Gau
    -0.61
     Bulgar
    -0.61
     Burg
    -0.61
    OWER
    -0.61
     Grizz
    -0.60
    POSITIVE LOGITS
     incidentally
    1.01
     preceded
    0.97
     famously
    0.93
    anwhile
    0.88
     admittedly
    0.86
     specialize
    0.85
     lacks
    0.84
     prominently
    0.84
     dominates
    0.83
     occupies
    0.82
    Activations Density 0.190%

    No Known Activations