INDEX

Explanations

phrases associated with reading

oai_token-act-pair · gpt-3.5-turbo

instances of the word "read" and its variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ortium

-0.64

iership

-0.60

vre

-0.59

penter

-0.58

 Launcher

-0.58

Ã±o

-0.57

uncture

-0.57

apolis

-0.56

udic

-0.55

 restart

-0.55

POSITIVE LOGITS

 aloud

1.32

just

1.02

dress

0.93

ahead

0.89

mitt

0.83

printed

0.81

written

0.81

\\\\\\\\

0.80

letter

0.78

 comprehension

0.76

Activations Density 0.026%

phrases associated with reading

instances of the word "read" and its variants

No Comments

No Known Activations