INDEX

Explanations

contractions where 't' is followed by an apostrophe and a verb

oai_token-act-pair · gpt-3.5-turbo

negative contractions and forms of the word "not."

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

cedes

-0.70

gra

-0.66

haul

-0.66

 scram

-0.64

leaders

-0.62

forms

-0.61

geist

-0.59

issance

-0.58

mares

-0.58

chairs

-0.58

POSITIVE LOGITS

ometimes

0.62

panic

0.62

 stray

0.61

otta

0.60

 Magazine

0.60

 Equality

0.59

ENSE

0.59

)?

0.58

Baz

0.57

chio

0.57

Activations Density 0.098%

contractions where 't' is followed by an apostrophe and a verb

negative contractions and forms of the word "not."

No Comments

No Known Activations