© 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. 10-RES-JB
    6. 2877
    Prev
    Next
    INDEX
    Explanations

    adjectives modified by '-est' or 'ful' which indicate a higher degree of a certain quality, such as the 'prettiest' or 'craziest'

    oai_token-act-pair · gpt-3.5-turbo

    descriptors related to amusing or bizarre experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Aram
    -0.61
     constitu
    -0.59
     Vi
    -0.58
     Domain
    -0.57
     alone
    -0.57
     Sag
    -0.57
     pregn
    -0.57
     Pax
    -0.56
     Indo
    -0.56
     Hak
    -0.56
    POSITIVE LOGITS
    iest
    2.03
    iness
    1.89
    ily
    1.74
    iful
    1.73
    ier
    1.71
    ied
    1.46
    ies
    1.42
    ifully
    1.42
    ying
    1.36
    eless
    1.32
    Activations Density 0.114%

    No Known Activations