INDEX

Explanations

phrases related to peace and peaceful actions

oai_token-act-pair · gpt-3.5-turbo

references to peaceful behavior or actions

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

MAC

-0.74

olls

-0.74

ails

-0.73

GPU

-0.70

ripp

-0.69

drivers

-0.69

Sales

-0.68

attr

-0.68

asper

-0.68

è¦ļéĨĴ

-0.67

POSITIVE LOGITS

 peaceful

0.97

peace

0.83

 minded

0.79

ness

0.78

\\\\\\\\

0.75

edIn

0.74

ysc

0.72

 resolution

0.71

edom

0.71

haven

0.70

Activations Density 0.015%

phrases related to peace and peaceful actions

references to peaceful behavior or actions

No Comments

No Known Activations