INDEX

Explanations

Words ending in 'o'

oai_token-act-pair · gpt-3.5-turbo

variations of the letter "o" and related linguistic structures

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

iolet

-0.80

arez

-0.76

vez

-0.74

oval

-0.70

ereo

-0.69

kowski

-0.69

OTT

-0.68

avin

-0.68

iris

-0.66

NL

-0.65

POSITIVE LOGITS

arial

0.81

ful

0.80

iness

0.75

naires

0.70

iest

0.69

istical

0.68

BOOK

0.68

 largeDownload

0.68

 motto

0.68

liest

0.67

Activations Density 0.232%

Words ending in 'o'

variations of the letter "o" and related linguistic structures

No Comments

No Known Activations