INDEX

Explanations

proper nouns or names

oai_token-act-pair · gpt-3.5-turbo

the occurrences of the word "like"

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

 ILCS

-0.90

istry

-0.85

istics

-0.70

mable

-0.70

istically

-0.69

aries

-0.68

ONT

-0.67

imester

-0.67

rooms

-0.66

ogyn

-0.64

POSITIVE LOGITS

bike

0.99

lihood

0.93

mens

0.87

jriwal

0.87

gged

0.84

ernel

0.81

llan

0.81

yu

0.73

ye

0.71

manship

0.71

Activations Density 0.018%

proper nouns or names

the occurrences of the word "like"

No Comments

No Known Activations