© 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. 8294
    Prev
    Next
    INDEX
    Explanations

    references to negative outcomes or quantitative losses

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "losses" and related terms

    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
    Created
    -0.71
    pol
    -0.66
    dayName
    -0.66
    JB
    -0.64
    Fram
    -0.63
    cart
    -0.63
    ISTER
    -0.63
    Offic
    -0.62
    bara
    -0.61
    pter
    -0.61
    POSITIVE LOGITS
     losses
    3.83
     loss
    2.33
    loss
    2.18
     Loss
    2.13
     defeats
    1.90
     loses
    1.74
     setbacks
    1.67
     losers
    1.65
     victories
    1.58
     failures
    1.55
    Activations Density 0.014%

    No Known Activations