© 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. 0-RES-JB
    11. 10353
    Prev
    Next
    INDEX
    Explanations

    terms related to security breaches or unauthorized access

    oai_token-act-pair · gpt-3.5-turbo

    terms related to junk or unwanted elements, as well as notions of being hijacked or taken over

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     bol
    -0.86
     fir
    -0.84
     bir
    -0.83
     latitude
    -0.75
     brace
    -0.74
    achus
    -0.70
    atal
    -0.70
     acc
    -0.70
     clamp
    -0.70
    BP
    -0.70
    POSITIVE LOGITS
     Junk
    3.02
    otive
    1.47
     hijacked
    1.36
     Mafia
    1.17
     Marketing
    1.09
     unheard
    1.07
    geist
    1.05
     Annie
    1.05
     Riot
    1.00
     Synd
    0.99
    Activations Density 0.052%

    No Known Activations