© 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. 1-RES-JB
    11. 5499
    Prev
    Next
    INDEX
    Explanations

    phrases indicating absence or insufficiency

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "lack" conveying absence or deficiency

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Drac
    -0.66
     bath
    -0.64
     massage
    -0.63
     Razor
    -0.61
     Circus
    -0.60
     precincts
    -0.59
     Amen
    -0.58
    iph
    -0.57
     Mafia
    -0.57
    eor
    -0.57
    POSITIVE LOGITS
    lust
    1.05
    luster
    0.90
    lessly
    0.84
     thereof
    0.82
     inconsistency
    0.75
    luck
    0.75
    ¼
    0.74
    nces
    0.73
    enz
    0.73
     miser
    0.73
    Activations Density 0.015%

    No Known Activations