© 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. 9132
    Prev
    Next
    INDEX
    Explanations

    specific mention of quantities or statistically significant figures

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "alone" indicating emphasis on isolation or singularity

    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
    aptic
    -0.79
    anty
    -0.77
    emies
    -0.69
    eways
    -0.68
    otech
    -0.67
    inflamm
    -0.66
    olid
    -0.65
    nit
    -0.65
    emy
    -0.65
     Briggs
    -0.65
    POSITIVE LOGITS
     alone
    0.80
    displayText
    0.67
    alone
    0.66
     unaccompanied
    0.65
     Alone
    0.65
     suffice
    0.65
    SPONSORED
    0.65
    Ident
    0.64
    =~=~
    0.62
     admissions
    0.61
    Activations Density 0.012%

    No Known Activations