INDEX

Explanations

phrases related to negative societal events or incidents

oai_token-act-pair · gpt-3.5-turbo

references to violent incidents and their prevalence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ãĤ£

-0.71

ITS

-0.67

anu

-0.66

AMA

-0.64

Minecraft

-0.64

#$

-0.63

dit

-0.61

ovo

-0.61

amphetamine

-0.60

MAP

-0.60

POSITIVE LOGITS

 prolifer

1.21

 abound

1.20

 commonplace

1.19

 occurring

1.12

 popping

1.09

 occur

1.08

 frequent

1.06

 sporadic

1.04

 plentiful

1.00

 spor

0.97

Activations Density 0.575%

phrases related to negative societal events or incidents

references to violent incidents and their prevalence

No Comments

No Known Activations