INDEX

Explanations

words related to changes or modifications

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "changed" and its variations, indicating modifications or alterations

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

ebook

-0.68

emp

-0.65

Ga

-0.63

 Cure

-0.63

Song

-0.63

 Nile

-0.62

ace

-0.62

Found

-0.61

 Apostle

-0.60

Ca

-0.60

POSITIVE LOGITS

terday

1.00

 drastically

0.84

 accordingly

0.83

 chronological

0.79

 dramatically

0.76

 sharply

0.75

 gears

0.74

otom

0.74

 abruptly

0.74

 administr

0.72

Activations Density 0.145%

words related to changes or modifications

occurrences of the word "changed" and its variations, indicating modifications or alterations

No Comments

No Known Activations