© 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. 22283
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a comparison or competition between entities

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "vs" indicating a comparison or matchup

    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
    appropriately
    -0.71
    ANY
    -0.66
    rouse
    -0.64
    iries
    -0.64
    erity
    -0.62
    ©¶æ¥µ
    -0.60
    ¿
    -0.59
     discreet
    -0.58
     Belg
    -0.58
    oho
    -0.57
    POSITIVE LOGITS
     vs
    3.76
     versus
    3.09
     Vs
    2.72
     Versus
    2.08
    vs
    2.05
     VS
    1.87
     compared
    1.23
     against
    1.13
    Vs
    1.08
    VS
    0.99
    Activations Density 0.013%

    No Known Activations