INDEX

Explanations

phrases related to historical information, particularly about events or facts

oai_token-act-pair · gpt-3.5-turbo

references to historical events or records

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

umbers

-0.75

emouth

-0.73

atron

-0.71

uters

-0.70

ople

-0.68

uple

-0.68

aband

-0.67

tackle

-0.65

hement

-0.65

cliffe

-0.64

POSITIVE LOGITS

 Origins

0.87

Background

0.84

========

0.84

 origins

0.82

Early

0.81

Past

0.80

Orig

0.78

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

0.77

Spoiler

0.77

Appearance

0.71

Activations Density 0.086%

phrases related to historical information, particularly about events or facts

references to historical events or records

No Comments

No Known Activations