© 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. 848
    Prev
    Next
    INDEX
    Explanations

    references to a sports team named "Burrows" or related terms

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "rows" and its variations, indicating a focus on structures or lists

    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
    ãĥ¼ãĥĨ
    -0.68
     PLAN
    -0.66
    piece
    -0.65
     circumstance
    -0.64
     burden
    -0.64
     hemor
    -0.61
     dearly
    -0.60
     reinforcement
    -0.59
     modeling
    -0.58
     incentive
    -0.58
    POSITIVE LOGITS
    icz
    1.12
    rows
    1.10
    restling
    1.09
    orld
    0.89
    ski
    0.88
    hip
    0.88
    dropping
    0.87
    ight
    0.86
    heed
    0.86
    uggest
    0.85
    Activations Density 0.007%

    No Known Activations