INDEX

Explanations

adjectives with neutral or negative connotations

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "but."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

icipated

-0.75

rongh

-0.72

oire

-0.72

inas

-0.72

zag

-0.71

unk

-0.70

folio

-0.70

unks

-0.70

onica

-0.69

tnc

-0.69

POSITIVE LOGITS

 nonetheless

1.35

 nevertheless

1.26

hey

1.17

 alas

1.14

 fortunately

1.06

 luckily

1.04

 certainly

1.01

 beware

0.95

tons

0.92

 definitely

0.92

Activations Density 0.173%

adjectives with neutral or negative connotations

phrases that include the word "but."

No Comments

No Known Activations