INDEX

Explanations

references to events or behaviors that occurred in the past

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "past."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Franch

-0.75

wagon

-0.71

hat

-0.66

oola

-0.65

utters

-0.60

atism

-0.60

NEY

-0.59

rapnel

-0.59

BuyableInstoreAndOnline

-0.59

ithmetic

-0.59

POSITIVE LOGITS

ebin

1.42

 tense

1.11

decade

0.99

 week

0.79

 month

0.78

iche

0.77

ĸļ

0.75

orate

0.73

 semester

0.73

few

0.72

Activations Density 0.023%

references to events or behaviors that occurred in the past

occurrences of the word "past."

No Comments

No Known Activations