© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 641
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the overall situation or concept being discussed

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing the concept of "the whole thing."

    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
     imminent
    -0.65
     occasional
    -0.60
    odan
    -0.59
     brakes
    -0.59
     Cutter
    -0.59
     heels
    -0.58
    ello
    -0.58
    Soul
    -0.58
    eka
    -0.58
     mM
    -0.58
    POSITIVE LOGITS
     laun
    0.80
     except
    0.80
    except
    0.79
    yss
    0.76
     revolves
    0.74
    代
    0.72
    ãĤ¼ãĤ¦ãĤ¹
    0.71
     shenan
    0.70
    anmar
    0.69
     Including
    0.69
    Activations Density 0.161%

    No Known Activations