INDEX

Explanations

mentions of specific named products or entities

oai_token-act-pair · gpt-3.5-turbo

specific products or notable statements regarding significant topics

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 HuffPost

-0.54

rosso

-0.53

 explanations

-0.51

choes

-0.51

genre

-0.51

 Yiannopoulos

-0.50

utics

-0.49

enses

-0.49

 wary

-0.48

 DevOnline

-0.48

POSITIVE LOGITS

 ..."

1.05

 negro

0.76

..."

0.74

..

0.74

 alot

0.72

 ....

0.71

 ONLY

0.69

 americ

0.68

!!

0.67

 âĢ¦"

0.66

Activations Density 1.949%

mentions of specific named products or entities

specific products or notable statements regarding significant topics

No Comments

No Known Activations