© 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. 0-RES-JB
    11. 14860
    Prev
    Next
    INDEX
    Explanations

    instances where the word "fifth" appears in a text

    oai_token-act-pair · gpt-3.5-turbo

    references to ordinal rankings, particularly the terms "fifth" and "sixth."

    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
    storm
    -0.61
    ache
    -0.60
    amo
    -0.60
    RT
    -0.60
     Restore
    -0.59
    etting
    -0.58
    ite
    -0.58
     Allows
    -0.56
     Recon
    -0.56
    irtual
    -0.56
    POSITIVE LOGITS
     fifth
    3.45
     sixth
    3.02
     seventh
    2.94
     fourth
    2.66
     eighth
    2.64
     ninth
    2.57
    fifth
    2.46
     tenth
    2.40
     third
    2.36
    fourth
    1.99
    Activations Density 0.018%

    No Known Activations