INDEX

Explanations

HTML classes within text

oai_token-act-pair · gpt-3.5-turbo

references to specific classifications or categories, particularly related to "class."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 stump

-0.73

edia

-0.68

 spur

-0.66

²¾

-0.65

 hypot

-0.63

vernment

-0.63

wit

-0.60

wards

-0.58

issance

-0.58

 Senegal

-0.57

POSITIVE LOGITS

classes

1.16

ifications

1.09

ifiers

1.08

ifier

1.04

class

1.03

 class

1.01

Class

0.99

 classes

0.99

ifies

0.93

ifying

0.93

Activations Density 0.023%

HTML classes within text

references to specific classifications or categories, particularly related to "class."

No Comments

No Known Activations