INDEX

Explanations

references to things being among the largest or most significant in a particular category

oai_token-act-pair · gpt-3.5-turbo

references to significant or major entities/events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

iture

-0.93

 thereof

-0.77

ESSION

-0.74

aeus

-0.73

fulness

-0.68

ume

-0.68

Zup

-0.67

é¾

-0.67

ukong

-0.67

Rex

-0.66

POSITIVE LOGITS

 oldest

1.27

 hottest

1.27

 earliest

1.25

 coolest

1.22

 largest

1.17

 busiest

1.17

 strongest

1.17

 heaviest

1.14

 fastest

1.13

 finest

1.11

Activations Density 0.080%

references to things being among the largest or most significant in a particular category

references to significant or major entities/events

No Comments

No Known Activations