INDEX

Explanations

references to specific images or captions in a news context

oai_token-act-pair · gpt-3.5-turbo

references to specific images or visual content in documents

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

pill

-0.79

ocr

-0.68

hop

-0.68

boro

-0.67

 Norn

-0.67

SG

-0.65

Sword

-0.65

blank

-0.64

Â·Â·

-0.63

lethal

-0.62

POSITIVE LOGITS

 Photos

0.88

window

0.87

 Tanks

0.73

 Caption

0.69

 WATCHED

0.69

kay

0.65

 FILE

0.63

 msec

0.61

 captures

0.60

 ammon

0.59

Activations Density 0.063%

references to specific images or captions in a news context

references to specific images or visual content in documents

No Comments

No Known Activations