INDEX

Explanations

the name "Doc" preceded by a number indicating different levels of importance or significance

oai_token-act-pair · gpt-3.5-turbo

references to "Doc" or similar terms related to individuals with the title "Doc."

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

theless

-0.72

places

-0.68

Interstitial

-0.65

Rated

-0.64

WAYS

-0.64

 perme

-0.62

 susceptibility

-0.60

 indignation

-0.60

 Mechdragon

-0.59

mare

-0.59

POSITIVE LOGITS

herty

1.17

uments

1.15

ilic

0.98

ually

0.97

Doc

0.96

umen

0.94

ilt

0.94

ocular

0.92

ents

0.92

iencies

0.92

Activations Density 0.018%

the name "Doc" preceded by a number indicating different levels of importance or significance

references to "Doc" or similar terms related to individuals with the title "Doc."

No Comments

No Known Activations