INDEX

Explanations

adverbs ending in 'ly'

oai_token-act-pair · gpt-3.5-turbo

adverbs that describe extreme or exaggerated qualities

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 GOODMAN

-0.79

erto

-0.68

 Seasons

-0.68

 doorway

-0.66

raltar

-0.65

 brokers

-0.65

 commissioners

-0.64

ilater

-0.62

eport

-0.62

arta

-0.61

POSITIVE LOGITS

BuyableInstoreAndOnline

0.86

 impressed

0.75

fal

0.73

 positioned

0.69

named

0.69

present

0.69

 LIMITED

0.68

nit

0.66

 priced

0.66

ãĤ¨

0.65

Activations Density 0.044%

adverbs ending in 'ly'

adverbs that describe extreme or exaggerated qualities

No Comments

No Known Activations