© 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. 22033
    Prev
    Next
    INDEX
    Explanations

    phrases expressing disappointment or sadness

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "that" indicating expressions of regret or disappointment

    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
    •layer 4 - sadness

    No Comments

    Negative Logits
    orah
    -0.69
    ead
    -0.60
    MH
    -0.60
    ocaust
    -0.60
    istance
    -0.59
    thur
    -0.58
    zman
    -0.58
    utable
    -0.58
    ione
    -0.57
    apor
    -0.57
    POSITIVE LOGITS
    soever
    0.76
     surrounds
    0.75
    cher
    0.73
    ndra
    0.67
    fy
    0.65
    76561
    0.64
    terday
    0.63
    ovie
    0.63
     they
    0.63
     pesky
    0.61
    Activations Density 0.268%

    No Known Activations