INDEX

Explanations

phrases related to legal and regulatory language

oai_token-act-pair · gpt-3.5-turbo

symbols and punctuation associated with expressions of opinion

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

wagen

-0.72

 conduc

-0.71

 destro

-0.69

creen

-0.68

 mosqu

-0.68

terday

-0.67

 Drawn

-0.66

 Dupl

-0.65

 grips

-0.64

 Belg

-0.64

POSITIVE LOGITS

ª

1.01

most

1.00

Ĵ

0.94

¹

0.94

ĳ

0.89

ł

0.89

actual

0.88

¼

0.87

option

0.87

race

0.87

Activations Density 0.078%

phrases related to legal and regulatory language

symbols and punctuation associated with expressions of opinion

No Comments

No Known Activations