© 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. 10-RES-JB
    11. 20410
    Prev
    Next
    INDEX
    Explanations

    phrases related to collaboration and project management

    oai_token-act-pair · gpt-3.5-turbo

    terms related to stakeholder engagement and collaborative processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     faked
    -0.79
     unlucky
    -0.78
     booze
    -0.77
     mysteriously
    -0.76
     fried
    -0.75
     stupidity
    -0.74
     prank
    -0.74
     innocence
    -0.74
     Mour
    -0.74
     cursing
    -0.73
    POSITIVE LOGITS
     Implementation
    1.15
     stakeholders
    1.14
    Recommend
    1.12
     priorit
    1.09
     initiatives
    1.09
     practicable
    1.08
     Framework
    1.07
     recommendations
    1.06
     Recommend
    1.05
     objectives
    1.04
    Activations Density 0.696%

    No Known Activations