INDEX

Explanations

comparative adjectives indicating size or scale

oai_token-act-pair · gpt-3.5-turbo

the word "bigger" in various contexts

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

shire

-0.90

endant

-0.70

endants

-0.70

onso

-0.69

pta

-0.69

 Michaels

-0.67

ices

-0.67

chron

-0.67

ION

-0.66

syn

-0.66

POSITIVE LOGITS

 than

0.99

 Than

0.90

picture

0.90

oted

0.85

than

0.85

 eleph

0.83

 bang

0.82

 scale

0.81

 stature

0.80

 picture

0.80

Activations Density 0.015%

comparative adjectives indicating size or scale

the word "bigger" in various contexts

No Comments

No Known Activations