© 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. 4-RES-JB
    11. 7588
    Prev
    Next
    INDEX
    Explanations

    the verb "is" followed by an adjective or noun phrase

    oai_token-act-pair · gpt-3.5-turbo

    expressions indicating existence or state

    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
     Hitch
    -0.66
     Poll
    -0.64
    itches
    -0.63
     Lack
    -0.62
    heres
    -0.61
    Footnote
    -0.60
     mayors
    -0.59
     Peoples
    -0.59
    idence
    -0.58
     Chev
    -0.58
    POSITIVE LOGITS
     meant
    0.88
     supposed
    0.86
     happening
    0.81
    actually
    0.81
     transpired
    0.80
    really
    0.78
    omorphic
    0.75
     presently
    0.73
     termed
    0.72
     actually
    0.72
    Activations Density 0.110%

    No Known Activations