INDEX

Explanations

phrases that express strong opinions or assertions

oai_token-act-pair · gpt-3.5-turbo

phrases that express comparisons or examples

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

abase

-0.74

ertain

-0.72

OVA

-0.71

ennes

-0.70

ongo

-0.69

MAS

-0.68

ktop

-0.67

vt

-0.65

ecast

-0.62

HT

-0.62

POSITIVE LOGITS

"'

1.26

"@

1.24

"#

1.19

"(

1.17

 "âĢ¦

1.17

"[

1.16

 "...

1.14

"$

1.07

1.06

"<

0.99

Activations Density 0.255%

phrases that express strong opinions or assertions

phrases that express comparisons or examples

No Comments

No Known Activations