INDEX

Explanations

phrases related to a specific object or concept mentioned earlier in the text

oai_token-act-pair · gpt-3.5-turbo

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

icons

-0.77

english

-0.72

orians

-0.71

ormons

-0.71

izons

-0.70

ocks

-0.69

å§«

-0.69

apolis

-0.69

anyahu

-0.69

osponsors

-0.68

POSITIVE LOGITS

 fateful

1.29

 particular

1.27

 same

1.22

 pesky

0.97

cher

0.91

 timeframe

0.90

 portion

0.88

 exact

0.87

ched

0.87

 subset

0.85

Activations Density 0.121%

phrases related to a specific object or concept mentioned earlier in the text

No Comments

No Known Activations