INDEX

Explanations

Chinese characters and other unidentifiable symbols

oai_token-act-pair · gpt-3.5-turbo

specific names or terms related to places and institutions

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

imeters

-0.64

âķĲâķĲ

-0.56

 sand

-0.56

GREEN

-0.52

 centr

-0.51

restling

-0.51

 baths

-0.50

Cu

-0.50

 quartz

-0.49

isner

-0.49

POSITIVE LOGITS

ĪĴ

0.74

 Ribbon

0.68

alion

0.60

¥µ

0.60

illard

0.59

eele

0.58

Ö¼

0.56

 Trials

0.54

 Despair

0.52

Ĭ±

0.52

Activations Density 1.328%

Chinese characters and other unidentifiable symbols

specific names or terms related to places and institutions

No Comments

No Known Activations