© 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. 4-RES-JB
    11. 382
    Prev
    Next
    INDEX
    Explanations

    the word "terrible."

    oai_token-act-pair · gpt-3.5-turbo

    expressions of negative opinions or criticisms

    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
    Not in Any Lists

    No Comments

    Negative Logits
    ership
    -0.96
    pai
    -0.88
    ilus
    -0.87
    irs
    -0.85
    ovember
    -0.78
    cript
    -0.78
    ersen
    -0.77
    eters
    -0.76
    aver
    -0.76
    illance
    -0.76
    POSITIVE LOGITS
     havoc
    0.86
     nightmares
    0.85
     sounding
    0.83
     headache
    0.83
     awful
    0.80
     nightmare
    0.79
     horrible
    0.79
     adolesc
    0.78
     smelling
    0.75
     karma
    0.74
    Activations Density 0.013%

    No Known Activations