© 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. 4-RES-JB
    11. 12405
    Prev
    Next
    INDEX
    Explanations

    the word 'bare' with either body parts or concepts

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of "bare" or minimal conditions and essentials

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

    No Comments

    Negative Logits
    anwhile
    -0.83
    å°Ĩ
    -0.78
    alez
    -0.76
    oldemort
    -0.74
    abulary
    -0.73
    ickr
    -0.70
    agents
    -0.70
     Democr
    -0.70
    afort
    -0.69
    utra
    -0.69
    POSITIVE LOGITS
    footed
    1.12
    faced
    1.08
    bones
    1.06
    foot
    1.05
    clad
    0.96
     bare
    0.95
    handed
    0.91
    feet
    0.91
    chest
    0.89
    hide
    0.86
    Activations Density 0.009%

    No Known Activations