INDEX

Explanations

phrases containing the word "and" as a connecting element

oai_token-act-pair · gpt-3.5-turbo

the conjunction "and"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Powered

-0.70

floor

-0.69

Street

-0.68

rug

-0.66

SPA

-0.63

uction

-0.63

grad

-0.62

scribe

-0.62

odcast

-0.61

irc

-0.61

POSITIVE LOGITS

 consequently

1.12

 therefore

1.11

 thus

1.04

 hence

1.02

 then

1.00

rogens

1.00

rogen

0.99

 possibly

0.95

 THEN

0.95

 secondly

0.94

Activations Density 0.470%

phrases containing the word "and" as a connecting element

the conjunction "and"

No Comments

No Known Activations