INDEX

Explanations

references to large quantities of physical items, specifically 'tons'

oai_token-act-pair · gpt-3.5-turbo

quantitative measures of weight, specifically in tons or tonnes

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

gotten

-0.70

 resemb

-0.67

hall

-0.67

Dub

-0.66

Loop

-0.63

cards

-0.63

Journal

-0.63

WHERE

-0.62

govern

-0.62

 Advice

-0.62

POSITIVE LOGITS

lbs

1.01

 tonnes

0.97

 barrels

0.94

 pounds

0.93

poons

0.91

 tons

0.86

 apiece

0.82

atsu

0.77

iles

0.74

 gallons

0.74

Activations Density 0.016%

references to large quantities of physical items, specifically 'tons'

quantitative measures of weight, specifically in tons or tonnes

No Comments

No Known Activations