© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 18133
    Prev
    Next
    INDEX
    Explanations

    words related to transactions or instructions

    oai_token-act-pair · gpt-3.5-turbo

    strong indicators of legislative terms and references

    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
    cephal
    -0.73
    Fax
    -0.70
    Cast
    -0.70
    erb
    -0.66
    iste
    -0.66
     Ne
    -0.65
    Lens
    -0.65
    Corp
    -0.63
    Alias
    -0.63
    Jet
    -0.63
    POSITIVE LOGITS
     HIGH
    2.05
     WORK
    1.99
     YEAR
    1.99
     PROGRAM
    1.98
     INTER
    1.96
     LAST
    1.96
     COUN
    1.95
     SHAR
    1.93
     MIN
    1.93
     STAND
    1.93
    Activations Density 0.101%

    No Known Activations