© 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. 8-RES-JB
    11. 7595
    Prev
    Next
    INDEX
    Explanations

    instances of actions related to deleting information

    oai_token-act-pair · gpt-3.5-turbo

    references to the act of deleting items or accounts in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    annis
    -0.85
    ebus
    -0.74
    acs
    -0.72
    Building
    -0.68
    verning
    -0.68
     negotiators
    -0.67
    ETF
    -0.66
    croft
    -0.64
    orsi
    -0.63
    gio
    -0.63
    POSITIVE LOGITS
     deleted
    0.87
     Delete
    0.86
     delet
    0.83
    leted
    0.79
     delete
    0.76
    itor
    0.75
    username
    0.69
    aneous
    0.69
    ution
    0.68
     unnecessary
    0.67
    Activations Density 0.040%

    No Known Activations