© 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. 2-RES-JB
    11. 18732
    Prev
    Next
    INDEX
    Explanations

    mentions of specific names, like "Megan" and "Randall"

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the name "Megan" and references to the character's relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    lesiastical
    -0.97
    estinal
    -0.88
    akespe
    -0.88
    achy
    -0.85
    lasses
    -0.82
    rosis
    -0.82
    yer
    -0.80
    hid
    -0.80
    esthes
    -0.80
    anamo
    -0.79
    POSITIVE LOGITS
     Randall
    1.01
     Cue
    0.95
     Danger
    0.86
     Goods
    0.85
     Pearce
    0.83
     Gale
    0.80
     Haram
    0.79
     Megan
    0.78
     Mun
    0.77
     Mew
    0.77
    Activations Density 0.022%

    No Known Activations