INDEX

Explanations

numbers from captions or descriptions in a set of photos

oai_token-act-pair · gpt-3.5-turbo

numerical data or statistics related to images or reports

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

romeda

-0.69

 conversion

-0.68

olate

-0.64

isoft

-0.64

burning

-0.61

istic

-0.61

ilic

-0.60

ieu

-0.59

 Ukrain

-0.59

spread

-0.58

POSITIVE LOGITS

 IMAGES

0.78

ECA

0.77

 Â·

0.73

 SHARES

0.71

 Ways

0.70

 VIDEOS

0.70

Zup

0.69

0.68

Activations Density 0.130%

numbers from captions or descriptions in a set of photos

numerical data or statistics related to images or reports

No Comments

No Known Activations