© 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. 12015
    Prev
    Next
    INDEX
    Explanations

    N/A - Neuron 4 does not appear to have consistent activations related to a specific type of content

    oai_token-act-pair · gpt-3.5-turbo

    a specific character or sequence of characters that may indicate formatting or encoding anomalies

    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
     Franch
    -0.84
     order
    -0.79
     Liqu
    -0.72
     rehearsal
    -0.66
     Reese
    -0.66
     paperwork
    -0.64
     job
    -0.64
     Furn
    -0.64
    fin
    -0.64
     recogn
    -0.63
    POSITIVE LOGITS
    ³
    4.35
    ¹
    2.02
    ¿
    2.01
    ²
    2.00
    ¶
    2.00
    ±
    1.97
    ½
    1.92
    µ
    1.85
    ª
    1.81
    »
    1.81
    Activations Density 0.008%

    No Known Activations