© 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. 9-RES-JB
    11. 21172
    Prev
    Next
    INDEX
    Explanations

    references to rumors or speculations

    oai_token-act-pair · gpt-3.5-turbo

    references to rumors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     humans
    -0.67
     Cornell
    -0.63
     electronically
    -0.62
     learning
    -0.62
     Levi
    -0.61
     spaces
    -0.60
    lihood
    -0.60
     Syri
    -0.60
     Cassidy
    -0.60
     human
    -0.59
    POSITIVE LOGITS
    pled
    1.41
    ble
    1.22
    inating
    1.20
    atis
    1.16
    oured
    1.13
    inatory
    1.07
    pling
    1.01
    ination
    1.00
    ours
    0.98
    ptions
    0.97
    Activations Density 0.040%

    No Known Activations