INDEX

Explanations

mentions of the word "water"

oai_token-act-pair · gpt-3.5-turbo

references to water and its related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Reloaded

-0.92

ablishment

-0.87

ures

-0.81

eering

-0.80

eways

-0.72

ificant

-0.71

urities

-0.66

doms

-0.66

urations

-0.66

eki

-0.65

POSITIVE LOGITS

melon

1.63

colour

1.12

proof

0.98

loo

0.97

marked

0.92

 fluor

0.92

color

0.90

tight

0.90

falls

0.89

 buffalo

0.89

Activations Density 0.024%

mentions of the word "water"

references to water and its related contexts

No Comments

No Known Activations