INDEX

Explanations

words related to taking risks or making significant choices

oai_token-act-pair · gpt-3.5-turbo

terms related to making choices and taking risks

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

force

-0.74

GW

-0.73

winner

-0.72

è¦ļéĨĴ

-0.68

gon

-0.68

raq

-0.68

iation

-0.66

BILL

-0.65

dy

-0.63

akh

-0.63

POSITIVE LOGITS

 abandonment

0.74

IPM

0.73

 captivity

0.73

 relinqu

0.73

 Braun

0.72

 shortcuts

0.71

 virginity

0.70

 hiber

0.68

 discretion

0.68

 anonymity

0.66

Activations Density 0.124%

words related to taking risks or making significant choices

terms related to making choices and taking risks

No Comments

No Known Activations