INDEX

Explanations

references to images in news articles

oai_token-act-pair · gpt-3.5-turbo

photo credits and image attributions

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

\":

-0.67

 defin

-0.63

ordinate

-0.59

rous

-0.57

 references

-0.57

 darling

-0.57

 word

-0.57

 aside

-0.57

 buggy

-0.57

 watering

-0.56

POSITIVE LOGITS

 Photography

0.88

photos

0.84

verett

0.82

 IMAGES

0.80

picture

0.77

ulton

0.75

Courtesy

0.75

Interstitial

0.74

allery

0.74

 Courtesy

0.73

Activations Density 0.175%

references to images in news articles

photo credits and image attributions

No Comments

No Known Activations