INDEX

Explanations

words related to hindrance or obstruction

oai_token-act-pair · gpt-3.5-turbo

phrases and words that indicate hindrances or barriers to progress

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

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Sham

-0.78

arse

-0.74

lat

-0.72

ocene

-0.71

worn

-0.70

olate

-0.67

oxic

-0.66

æ³

-0.66

arov

-0.65

iquette

-0.65

POSITIVE LOGITS

 progress

1.03

 accessing

0.99

 digestion

0.91

 access

0.86

 uptake

0.85

 development

0.83

 pursuit

0.81

 pursuing

0.80

 filming

0.79

 efforts

0.79

Activations Density 0.234%

words related to hindrance or obstruction

phrases and words that indicate hindrances or barriers to progress

No Comments

No Known Activations