© 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. 6-RES-JB
    11. 17352
    Prev
    Next
    INDEX
    Explanations

    numbers in the format of '5' followed by another number

    oai_token-act-pair · gpt-3.5-turbo

    markers indicating the end of text or section breaks

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    swick
    -0.64
    icol
    -0.64
     Hort
    -0.64
    worldly
    -0.60
    xual
    -0.60
    mia
    -0.60
    opter
    -0.59
     Hunts
    -0.59
     bapt
    -0.59
    perature
    -0.58
    POSITIVE LOGITS
    Thirty
    1.19
    âĺħ
    0.85
    th
    0.82
    010
    0.82
    678
    0.81
    0000
    0.77
    43
    0.77
    anging
    0.75
    pb
    0.74
     ILCS
    0.74
    Activations Density 0.115%

    No Known Activations