© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 6064
    Prev
    Next
    INDEX
    Explanations

    phrases indicating strong emotions or opinions, often in a negative context

    oai_token-act-pair · gpt-3.5-turbo

    instances of rants and violent outbursts

    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
    undai
    -0.88
    metics
    -0.86
    tarians
    -0.75
    ierrez
    -0.72
    rity
    -0.71
    cius
    -0.71
    liv
    -0.68
     Orchestra
    -0.66
     Assistance
    -0.66
    tarian
    -0.65
    POSITIVE LOGITS
     tir
    0.86
     vengeance
    0.84
    atical
    0.82
     against
    0.78
    quit
    0.76
     spree
    0.74
     rampage
    0.73
     AGA
    0.73
    ¯¯¯¯
    0.73
     rage
    0.71
    Activations Density 0.087%

    No Known Activations