© 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. 8-RES-JB
    11. 4225
    Prev
    Next
    INDEX
    Explanations

    expressions indicating beliefs or opinions

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating belief or confidence in outcomes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     disclosure
    -0.80
     disclosures
    -0.75
     testified
    -0.69
     deduction
    -0.68
     protested
    -0.67
     Occupations
    -0.67
     Brune
    -0.67
     mentions
    -0.67
     complains
    -0.67
     prohibitions
    -0.67
    POSITIVE LOGITS
     destined
    1.07
     ready
    1.06
     poised
    1.06
    Ready
    0.96
     viable
    0.95
     unbeat
    0.94
     unstoppable
    0.94
     primed
    0.91
     achievable
    0.91
    ready
    0.90
    Activations Density 0.654%

    No Known Activations