INDEX

Explanations

references to tables or structured content in text

oai_token-act-pair · gpt-3.5-turbo

mentions of tables of contents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

vernment

-0.85

 Directorate

-0.71

alez

-0.68

 Mehran

-0.67

ovich

-0.65

imal

-0.64

qua

-0.64

adobe

-0.63

 chancellor

-0.63

rily

-0.63

POSITIVE LOGITS

cloth

1.62

top

1.04

au

1.03

aux

0.99

tops

0.97

 scraps

0.96

 manners

0.93

poons

0.93

table

0.91

poon

0.87

Activations Density 0.026%

references to tables or structured content in text

mentions of tables of contents

No Comments

No Known Activations