INDEX

Explanations

phrases related to specific names or proper nouns, particularly "Ben" or variations of it

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names and titles

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

oslav

-0.79

FORMATION

-0.67

Interested

-0.65

 VIDE

-0.65

olor

-0.64

 veins

-0.64

ngth

-0.63

é¾įåĸļå£«

-0.63

glers

-0.62

 constitu

-0.61

POSITIVE LOGITS

rama

0.79

heid

0.72

gui

0.71

ilot

0.71

ente

0.70

aline

0.69

fen

0.67

ĳå£«

0.66

zeb

0.66

alion

0.66

Activations Density 0.071%

phrases related to specific names or proper nouns, particularly "Ben" or variations of it

proper nouns, specifically names and titles

No Comments

No Known Activations