INDEX

Explanations

phrases indicating emphasis or importance

oai_token-act-pair · gpt-3.5-turbo

the phrase "that," indicating emphasis on specific points or references in the text

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

apolis

-0.72

istics

-0.67

YS

-0.67

heid

-0.67

tones

-0.66

izont

-0.64

istor

-0.61

emis

-0.61

ington

-0.61

ophers

-0.61

POSITIVE LOGITS

 includes

1.02

 translates

0.99

 leads

0.95

 culminated

0.94

 happens

0.90

 entails

0.90

 constitutes

0.89

 resulted

0.88

 pesky

0.88

 manifests

0.87

Activations Density 0.118%

phrases indicating emphasis or importance

the phrase "that," indicating emphasis on specific points or references in the text

No Comments

No Known Activations