INDEX

Explanations

prepositions used to introduce a consequence or result

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize the impact or implications of an event on various subjects

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

querque

-0.85

Introduced

-0.77

cles

-0.69

zona

-0.67

ohn

-0.66

 laun

-0.64

Amb

-0.64

 Accessed

-0.64

portation

-0.62

ONSORED

-0.62

POSITIVE LOGITS

geries

1.09

bidden

1.08

gotten

1.03

gery

1.02

 awhile

0.99

 example

0.96

 instance

0.95

cing

0.93

 starters

0.92

 purposes

0.92

Activations Density 0.213%

prepositions used to introduce a consequence or result

phrases that emphasize the impact or implications of an event on various subjects

No Comments

No Known Activations