© 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. 11355
    Prev
    Next
    INDEX
    Explanations

    verbs or phrases indicating preference or support

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "favored" and its variants

    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
     apologise
    -0.79
     Rehab
    -0.78
     realise
    -0.73
    ember
    -0.69
    Act
    -0.69
    ecycle
    -0.68
    netflix
    -0.67
     analyse
    -0.66
    iry
    -0.64
    teness
    -0.64
    POSITIVE LOGITS
     favored
    3.85
     favoured
    3.06
     favors
    2.16
     favoring
    2.02
     favor
    1.89
     preferred
    1.81
    avored
    1.75
     favorites
    1.55
     favorable
    1.48
     prized
    1.42
    Activations Density 0.022%

    No Known Activations