INDEX

Explanations

words ending in 'es' with high activations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the suffix "es" at the end of words

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

è»

-0.74

iage

-0.73

ItemTracker

-0.70

Reviewer

-0.70

amental

-0.69

GAN

-0.66

allery

-0.66

=-=-=-=-

-0.65

staff

-0.65

ONSORED

-0.64

POSITIVE LOGITS

terday

1.27

ktop

1.22

peed

1.09

andro

1.08

earch

1.08

bians

1.06

pec

0.94

earchers

0.94

ury

0.90

leep

0.89

Activations Density 0.042%

words ending in 'es' with high activations

occurrences of the suffix "es" at the end of words

No Comments

No Known Activations