INDEX

Explanations

phrases where something is criticized or dismissed

oai_token-act-pair · gpt-3.5-turbo

instances of the word "as" in various contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

notice

-0.75

ppa

-0.71

ktop

-0.69

Posts

-0.67

OUS

-0.66

OTAL

-0.64

rone

-0.64

NET

-0.63

PsyNetMessage

-0.62

Atk

-0.61

POSITIVE LOGITS

ĪĴ

0.97

 well

0.94

 follows

0.91

 opposed

0.90

pires

0.86

piring

0.86

 soon

0.86

¥µ

0.83

phy

0.80

 evidenced

0.77

Activations Density 0.159%

phrases where something is criticized or dismissed

instances of the word "as" in various contexts

No Known Activations