INDEX

Explanations

quotes or statements made by people in interviews or news reports

oai_token-act-pair · gpt-3.5-turbo

instances of the word "told" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 utilized

-0.77

colored

-0.74

é¾įåĸļå£«

-0.70

 someday

-0.68

 colors

-0.68

 gotten

-0.64

inventory

-0.64

oros

-0.64

 utilizes

-0.63

 centered

-0.63

POSITIVE LOGITS

BBC

1.31

ITV

1.19

 Telegraph

1.14

 Fairfax

1.11

 Guardian

1.01

 broadcaster

0.98

 Mail

0.97

 Radio

0.96

Sky

0.95

 presenter

0.94

Activations Density 0.124%

quotes or statements made by people in interviews or news reports

instances of the word "told" in various contexts

No Comments

No Known Activations