© 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. 17705
    Prev
    Next
    INDEX
    Explanations

    multi-word phrases indicating a comparison or overview of something across different contexts or locations

    oai_token-act-pair · gpt-3.5-turbo

    the word "across" and its variations, often indicating studies or comparisons made in different contexts or regions

    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
    potion
    -0.75
    spot
    -0.73
    nery
    -0.72
    etic
    -0.66
    itious
    -0.65
    FORE
    -0.64
     bumped
    -0.62
    neum
    -0.61
    BALL
    -0.61
    zel
    -0.61
    POSITIVE LOGITS
     disciplines
    1.10
     continents
    0.97
    roads
    0.89
     generations
    0.85
    lines
    0.85
     genders
    0.82
     denominations
    0.80
     cultures
    0.79
     countries
    0.78
    states
    0.78
    Activations Density 0.039%

    No Known Activations