INDEX

Explanations

informative statements starting with "from the" or "the"

oai_token-act-pair · gpt-3.5-turbo

repeated phrases or references to a source, particularly focusing on context and attribution

oai_token-act-pair · gpt-4o-mini Triggered 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

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 eventually

-0.79

 having

-0.77

 repr

-0.76

 every

-0.73

 whatever

-0.73

 tending

-0.72

 cues

-0.72

 initially

-0.72

 deciding

-0.71

 requiring

-0.71

POSITIVE LOGITS

odore

1.31

 Editors

1.07

Score

1.07

 Authors

1.07

 Author

1.05

 author

1.04

atre

1.01

resa

0.95

Anti

0.93

 Editor

0.90

Activations Density 0.155%

informative statements starting with "from the" or "the"

repeated phrases or references to a source, particularly focusing on context and attribution

No Comments

No Known Activations