INDEX

Explanations

terms and phrases related to being referred to or named

oai_token-act-pair · gpt-3.5-turbo

instances of the word "called" referring to definitions or labels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

edia

-0.77

bilt

-0.74

SPONSORED

-0.66

olitics

-0.64

inth

-0.63

âĢ¢âĢ¢âĢ¢âĢ¢

-0.63

 isolate

-0.63

ajo

-0.62

LOD

-0.62

yip

-0.62

POSITIVE LOGITS

@#&

0.78

phas

0.62

 forth

0.61

 upon

0.61

0.59

 attention

0.59

 Foss

0.59

ilee

0.59

Activations Density 0.048%

terms and phrases related to being referred to or named

instances of the word "called" referring to definitions or labels

No Comments

No Known Activations