INDEX

Explanations

images or persons that are being described within the text

oai_token-act-pair · gpt-3.5-turbo

instances of the word "pictured" along with associated visual representations or descriptions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mental

-0.89

ance

-0.81

angan

-0.79

ances

-0.76

osc

-0.76

forestation

-0.74

endo

-0.73

abilities

-0.73

alities

-0.72

ament

-0.72

POSITIVE LOGITS

 Thumbnails

0.95

 Buckingham

0.78

 pictured

0.78

 gallery

0.77

 Photographer

0.75

 Above

0.74

 IMAGES

0.74

Pict

0.73

=#

0.72

 Pict

0.72

Activations Density 0.018%

images or persons that are being described within the text

instances of the word "pictured" along with associated visual representations or descriptions

No Comments

No Known Activations