© 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. 0-RES-JB
    11. 23214
    Prev
    Next
    INDEX
    Explanations

    words related to specific characters or references that are not common in the English language

    oai_token-act-pair · gpt-3.5-turbo

    characters that appear multiple times in succession

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     commentary
    -0.67
     alert
    -0.66
     Reboot
    -0.66
     comment
    -0.64
     Tuesday
    -0.63
     Wednesday
    -0.61
     wo
    -0.60
     Sunday
    -0.60
     Ub
    -0.59
     Nich
    -0.59
    POSITIVE LOGITS
    Ī
    4.55
    Īè
    2.33
    ī
    2.13
    Ħ
    2.00
    Į
    1.91
    į
    1.84
    Ģ
    1.83
    ĥ
    1.82
    Ĩ
    1.80
    ĩ
    1.76
    Activations Density 0.006%

    No Known Activations