© 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. 20201
    Prev
    Next
    INDEX
    Explanations

    words related to physical or emotional states, such as 'emotions' or 'pain'

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the prefix "em-."

    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
     Lovecraft
    -0.83
    DragonMagazine
    -0.77
     Kavanaugh
    -0.75
     Draw
    -0.73
     Hitch
    -0.72
     Accord
    -0.71
     Woodward
    -0.70
     Kubrick
    -0.70
     Belt
    -0.69
     Wynne
    -0.68
    POSITIVE LOGITS
    igrants
    1.31
    igrated
    1.25
    igrant
    1.20
    igration
    1.14
    manuel
    1.10
    perors
    1.08
    blems
    1.08
    igrate
    1.06
    igr
    1.06
    bers
    1.05
    Activations Density 0.008%

    No Known Activations