INDEX

Explanations

keywords related to a specific day of the week, 'Wednesday'

oai_token-act-pair · gpt-3.5-turbo

occurrences of the day "Wednesday."

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

respons

-0.82

åĬ

-0.82

chest

-0.78

ebook

-0.77

abet

-0.77

pend

-0.77

heid

-0.72

ript

-0.70

audi

-0.69

Export

-0.69

POSITIVE LOGITS

 morning

1.58

 afternoon

1.51

 evening

1.42

 night

1.36

 mornings

1.20

morning

1.13

 nights

1.05

 evenings

1.05

 Night

1.05

 Morning

1.04

Activations Density 0.085%

keywords related to a specific day of the week, 'Wednesday'

occurrences of the day "Wednesday."

No Comments

No Known Activations