© 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. 2770
    Prev
    Next
    INDEX
    Explanations

    names and locations related to a specific figure

    oai_token-act-pair · gpt-3.5-turbo

    mentions of a specific geographical location, Rafah, and terms related to paralysis

    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
    ept
    -0.86
    yll
    -0.76
    eller
    -0.74
    esta
    -0.72
    XY
    -0.68
    dress
    -0.67
    ijk
    -0.67
    Redditor
    -0.66
    emo
    -0.66
     wee
    -0.66
    POSITIVE LOGITS
     paralyzed
    0.90
     Alley
    0.88
     paralysis
    0.84
    ulic
    0.81
    Downloadha
    0.80
    aded
    0.78
     paraly
    0.75
    uted
    0.75
    acs
    0.75
    utics
    0.72
    Activations Density 0.027%

    No Known Activations