© 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. 10867
    Prev
    Next
    INDEX
    Explanations

    names or terms related to the city "London" and variations of the word "abandon"

    oai_token-act-pair · gpt-3.5-turbo

    references to geographical locations or names

    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
    uid
    -0.78
     abund
    -0.73
     necessities
    -0.69
    Publisher
    -0.69
    rd
    -0.65
    adish
    -0.64
     sockets
    -0.63
     intest
    -0.63
     subsistence
    -0.62
    tf
    -0.61
    POSITIVE LOGITS
    neau
    1.04
    als
    0.97
    aline
    0.86
    nell
    0.85
     Donovan
    0.84
    nel
    0.80
     Winc
    0.78
    ment
    0.75
    ski
    0.74
    ovich
    0.74
    Activations Density 0.010%

    No Known Activations