INDEX

Explanations

mentions of real-world entities such as names of people, places, or organizations

oai_token-act-pair · gpt-3.5-turbo

abbreviations and acronyms related to geographical locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

âĶĢâĶĢ

-0.79

 Emirates

-0.73

Ø¯

-0.72

 STATS

-0.70

 Rockets

-0.68

 Cassandra

-0.68

AVG

-0.68

ARDS

-0.67

ARD

-0.66

Ø¨

-0.65

POSITIVE LOGITS

emonic

1.57

uchin

1.33

Mn

1.17

ument

1.01

uggets

1.01

ajo

1.00

uth

0.98

asonic

0.97

ason

0.96

ovember

0.96

Activations Density 0.011%

mentions of real-world entities such as names of people, places, or organizations

abbreviations and acronyms related to geographical locations

No Comments

No Known Activations