INDEX

Explanations

adjectives followed by a noun

oai_token-act-pair · gpt-3.5-turbo

phrases that express significant depth, intensity, or notable qualities 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

":[

-0.64

 Sheriff

-0.61

agara

-0.59

ivist

-0.59

 Trey

-0.58

isec

-0.57

sent

-0.57

 airs

-0.57

hester

-0.56

 locality

-0.55

POSITIVE LOGITS

warts

0.82

reated

0.81

enegger

0.75

ptin

0.70

itely

0.68

BAT

0.66

NetMessage

0.64

ainted

0.64

arently

0.63

>.

0.61

Activations Density 0.196%

adjectives followed by a noun

phrases that express significant depth, intensity, or notable qualities in various contexts

No Comments

No Known Activations