INDEX

Explanations

words related to corporate executives or high-ranking officials within a company

oai_token-act-pair · gpt-3.5-turbo

references to corporate executives

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

¬¼

-0.85

¼

-0.73

ruck

-0.70

INGTON

-0.70

lington

-0.69

DonaldTrump

-0.68

Temperature

-0.68

ttp

-0.67

TPS

-0.67

apo

-0.66

POSITIVE LOGITS

 executives

0.97

doms

0.84

 exec

0.84

 incent

0.73

otonin

0.70

 memo

0.69

 overseeing

0.69

 suites

0.68

stration

0.67

 executive

0.67

Activations Density 0.041%

words related to corporate executives or high-ranking officials within a company

references to corporate executives

No Comments

No Known Activations