© 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. 9370
    Prev
    Next
    INDEX
    Explanations

    unexpected, disruptive events or revelations

    oai_token-act-pair · gpt-3.5-turbo

    terms related to surprising or significant news events, particularly those using the word "bombshell."

    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
     Sabha
    -0.86
    anse
    -0.84
    asus
    -0.78
    unes
    -0.76
    apore
    -0.75
    emis
    -0.74
     Expend
    -0.71
    ouf
    -0.70
    alf
    -0.70
     Gladiator
    -0.70
    POSITIVE LOGITS
     bombshell
    0.80
     brim
    0.80
     fore
    0.73
    nuts
    0.71
     loose
    0.71
    heads
    0.70
    uckle
    0.69
     hail
    0.66
    head
    0.66
    theless
    0.66
    Activations Density 0.028%

    No Known Activations